Recurrent Aphthous Stomatitis Treatment Efficacy of Local Betamethasone Mouthwash and Systemic Colchicine Therapy and Their Effect on Serum and Salivary Cytokines

Home , Amlexanox, Betamethasone phosphate

RECURRENT APHTHOUS STOMATITIS TREATMENT EFFICACY OF LOCAL BETAMETHASONE MOUTHWASH AND SYSTEMIC COLCHICINE THERAPY AND THEIR EFFECT ON SERUM AND SALIVARY CYTOKINES

SURAB ALSAHAF B.D.S (Baghdad, Iraq) M.S.C (Eastman Dental Institute)

Thesis submitted in partial fulfillment for the Degree of Doctor of Philosophy in Oral Medicine of the University of London 2014

GKT Dental Institute Guy's Hospital King's College, London

This work is dedicated to the retirement of Professor SJ Challacombe and to the memory of Dr. Ron Wilson ABSTRACT

Background: Recurrent Aphthous Stomatitis (RAS) is a distinct oral mucosal disease of unknown aetiology. One of the factors involved in the pathogenesis of RAS may possibly be a cell mediated auto-immune response in which several cytokines appear to play a major role (Sun et al 2000). Hippocrates, the father of Medicine in 460-370 BC was the first one who used the term ‘Aphthae’ in relation to a disorder of the mouth, but the first clinical description appears to be that of Mikulicz and Kummel in 1898 (Sircus et al 1957). RAS is characterised by the spontaneous emergence of more than two bouts of oral ulcers per year and is not knowingly associated with ulcers anywhere else in the body or with underlying systemic diseases. Betamethasone mouthwash (Betnesol) as a topical steroid has become a standard therapy for RAS. In addition, Colchicine as a systemic therapy can be considered in severe cases, but to date there have been very few clinical trials reporting the efficacies for either.

Objectives: this randomised, prospective, parallel-group, single-centre, clinical trial over one year assessed the efficacy of Betnesol mouthwash (Betamethasone sodium phosphate 500mcg tablet dissolve in 10 ml of water and used as a mouthwash for 3 minutes QDS during ulcer attacks and BD in between) (Challacombe et al 1991) as monotherapy or as adjunct to Colchicine tablets 500mcg OD as systemic therapy compared with Colchicine tablets 500mcg OD as monotherapy. Furthermore, in order to enhance the understanding of the mode of action of the medication, nine specific cytokines in serum and saliva were assayed and related both to disease severity and to clinical changes as a result of treatment.

Method: 106 patients with RAS (Minor and Major) were randomized into 3 groups; 35 patients received Betnesol mouthwash QDS/BD, 35 patients received 500mcg Colchicine OD and 36 patients on Colchicine 0.5mg OD and Betnesol mouthwash QDS during ulcer attacks. An Ulcer Severity Score (USS) established in the department (Tappuni et al 2013) was used for

1 assessing the initial severity and monitoring any response to treatment. Responses were assessed at 3 monthly intervals for 12 months (a total of 5 visits). In addition, using the Luminex Multianalyte Profiling System (MAP), nine cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g) were measured in serum and salivary samples from 72 RAS patients (25 on Betnesol, 22 on Colchicine and 25 on Colchicine and Betnesol) and values compared with 34 healthy controls.

Results: 86 patients completed the full 12 months of the trial. The mean USS of the whole group before treatment was 34.9±7.3 (SD) and this decreased to 17.5±8.9 after 12 months therapy (p<0.001). The mean USS significantly reduced in each sub-group: Betnesol (34.6±7.6 to 15.7±9.3), Colchicine (32.5±6.9 to 17.7±9.7) and Colchicine plus Betnesol (36.9±6.7 to 19.2±7.8) (SD) (p<0.01) at 12 months, although the main effects were already apparent at six months. Significant reductions in the size, number, pain, sites affected and duration of ulcers and an increase in the length of ulcer-free periods (p<0.01) were found in all three groups. Salivary IL-8, IL-6, IL-10, IFN-g, and TNF-a as well as IL-8 in serum were found to be significantly raised in RAS patients compared with healthy controls (p<0.05). The mean salivary IL-6 and IFN-g levels were significantly greater when ulcers were present (p<0.05) and in Major RAS compared with Minor (p<0.05), whereas the other cytokine values appeared to be unrelated to the presence of ulcers or to the severity of RAS. Salivary IL-8, IL-6, IL-10 and TNF-a levels were decreased after 12 months treatment with Betnesol mouthwash (p<0.05) but not with Colchicine. In addition, levels of cytokines did not significantly correlate with the Ulcer Severity Score (USS).

Conclusion: This study demonstrates and confirms that Betnesol mouthwash is an effective therapy for RAS and affects all ulcer characteristics. Systemic Colchicine (500mcg) was also shown to be as effective as Betnesol. Combination of the two therapies was preferable for the management of Major RAS. In addition, Betnesol mouthwash resulted in a significant reduction of salivary pro-inflammatory cytokines levels in contrast to Colchicine. 2

ACKNOWLEDGEMENTS

I would like to express my sincerest gratitude to Professor SJ Challacombe for his remarkable support, encouragement, patience and continuous guidance throughout my period of study in the Department of Oral Medicine and preparation of this thesis. I am very grateful to Dr. A Tappuni for her continuous support and encouragement, who has supervised this research with her experience and knowledge. I would like to thank all the staff in the Department of Oral Medicine at Guy's Dental Institute for their support and it has been a privilege to work with them. I am indebted to many people for their help in the completion of this project, especially to Durdana Rahman in the Oral Pathology Department and to Professor A Lax, the director of graduate studies. Finally, many thanks to my family for their patience and support.

TABLE OF CONTENTS

ABSTRACT ...... 1 ACKNOWLEDGEMENTS ...... 3 TABLE OF CONTENTS ...... 4 LIST OF TABLES ...... 8 AIMS OF THIS STUDY ...... 10

CHAPTER ONE LITERATURE REVIEW RECURRENT APHTHOUS STOMATITIS

1.1 Introduction ...... 11 1.1.1 Gender and RAS ...... 15 1.1.2 Hormonal changes and RAS ...... 16 1.1.3 Stress and RAS ...... 16 1.2 Clinical Features ...... 19 1.2.1 Minor RAS ...... 19 1.2.2 Major RAS ...... 19 1.2.3 Herpetiform RAS ...... 20 1.2.4 Other classifications of RAS ...... 21 1.3 Genetic factors ...... 22 1.4 Immunopathogenesis ...... 26 1.4.1 Role of cytokines in RAS ...... 27 1.4.2 Peripheral Blood Mononuclear Cells (PBMCs) ...... 31 1.4.3 Lymphocyte sub-sets ...... 33 1.4.4 Polymorphonuclear (PMN) and Natural killer (NK) cells ...... 34 1.4.5 Behcet's Disease ...... 34 1.5 Methods of assessing Ulcer Severity ...... 36 1.5.1 Methods used in research studies ...... 36 1.6 Management of RAS ...... 39 1.6.1 Topical therapies ...... 40 1.6.2 Betamethasone sodium phosphate (Betnesol mouthwash) ...... 44

1.6.2.a Betamethasone and RAS ...... 45 1.6.2.b Betamethasone and cytokines ...... 46 1.6.3 Systemic Therapies ...... 48 1.6.4 Levamisole ...... 50 1.6.5 Dapsone ...... 51 1.6.6 Azathioprine ...... 52 1.6.7 Thalidomide ...... 53 1.6.8 Colchicine ...... 55 1.6.8.a Use of Colchicine in the treatment of RAS ...... 57 1.6.8.b Colchicine and its mode of action ...... 60 1.6.9 Other treatments of RAS ...... 62

CHAPTER 2 MATERIALS AND METHODS

2.1 Clinical Trial Design ...... 66 2.1.1 Inclusion criteria ...... 69 2.1.2 Exclusion criteria ...... 69 2.1.3 Randomisation ...... 70 2.1.4 Trial Medications ...... 71 2.1.5 Clinical study design ...... 72 2.1.6 Ulcer Severity Score (USS) ...... 75 2.1.7 Study Subjects ...... 78 2.1.8 Withdrawal of Subjects ...... 78 2.1.9 Assessment of Safety ...... 79 2.1.10 Assessment of Efficacy ...... 79 2.1.11 Method ...... 80 2.1.12 Patients Evaluation ...... 80 2.1.13 Trial Statistics ...... 82 2.2 Laboratory Study Design ...... 82 2.2.1 Serum samples ...... 83 2.2.2 Saliva samples ...... 83 2.2.3 Laboratory procedure ...... 84

2.2.4 Assay Procedure ...... 84 2.2.5 Luminex Multi-Analyte Profiling System (MAP) ...... 85 2.2.6 Reproducibility ...... 89

CHAPTER THREE RESULTS Comparison of three different modalities of treatments for the management of Recurrent Aphthous Stomatitis (RAS)

3.1 Complete series ...... 91 3.2 USS related to treatment ...... 95 3.2.1 (Group1) Betnesol mouthwash group ...... 95 3.2.1.a Ulcer Severity Scores (USS) before and after treatment with Betnesol mouthwash ...... 96 3.2.1.b Comparison of Ulcer Severity Scores (USS) in Minor and Major RAS after treatment with Betnesol mouthwash ...... 97 3.2.1.c Effect of Betnesol mouthwash treatment on six individual characteristics of ulcers ...... 102 3.2.2 (Group 2) Colchicine tablet group ...... 104 3.2.2.a Ulcer Severity Scores (USS) before and after systemic treatment with Colchicine ...... 105 3.2.2.b Comparison of Ulcer Severity Scores (USS) in Minor and Major RAS after systemic treatment with Colchicine ...... 106 3.2.2.c Effect of Colchicine treatment on six individual characteristics of ulcers ...... 111 3.2.3 (Group 3) Colchicine and Betnesol group ...... 113 3.2.3.a Ulcer Severity Scores (USS) before and after treatment with Colchicine and Betnesol ...... 114 3.2.3.b Comparison of Ulcer Severity Scores (USS) in Minor and Major RAS after treatment with Colchicine and Betnesol ...... 115 3.2.3.c Effect of Colchicine and Betnesol treatment on six individual characteristics of ulcers ...... 120 3.3 Comparison of Colchicine 0.5mg/day with 1mg/day in 10 Major RAS patients ...... 122

CHAPTER FOUR RESULTS Cytokines profile of Recurrent Aphthous Stomatitis (RAS) and the effects of treatments on serum and salivary cytokines

4.1 Introduction ...... 125 4.2 Cytokines in Healthy controls and RAS patients ...... 125 4.3 Cytokines in relation to the presence of ulcers ...... 132 4.4 Comparison of cytokines in Minor and Major RAS ...... 140 4.5 Effects of different modalities of treatments on Cytokines ...... 147

CHAPTER FIVE DISCUSSION Comparison of three different modalities of treatments for the management of Recurrent Aphthous Stomatitis (RAS)

5.1 Complete series ...... 156 5.2 Betnesol mouthwash group ...... 157 5.3 Colchicine tablet group ...... 160 5.4 Colchicine and Betnesol group ...... 164

CHAPTER SIX DISCUSSION Cytokines profile of Recurrent Aphthous Stomatitis (RAS) and the effects of treatments on serum and salivary cytokines

6.1 Cytokines in Healthy controls and RAS patients ...... 171 6.2 Cytokines in relation to the presence of ulcers ...... 176 6.3 Comparison of cytokines in Minor and Major RAS ...... 179 6.4 Effects of different modalities of treatments on Cytokines ...... 181

CONCLUSIONS OF THE STUDY ...... 186 FUTURE WORK ...... 187 REFERENCES ...... 188

LIST OF TABLES

Table 1 Prevalence of RAS

Table 2 Stress and RAS

Table 3 Genetics and RAS

Table 4 Cytokines and RAS

Table 5 Topical treatment of RAS

Table 6 Systemic treatment of RAS

Table 7 Other types of RAS treatment

Table 8 Inter-Reproducibility of nine cytokines (saliva)

Table 9 Inter-Reproducibility of nine cytokines (serum)

Table 10 Intra-Reproducibility of nine cytokines (saliva)

Table 11 Intra-Reproducibility of nine cytokines (serum)

Table 12 Summary of Ulcer Severity Scores (USS) before and after treatment with Betnesol mouthwash QDS/BD for 12 months in patients with Minor and Major RAS

Table 13 Summary of ulcer characteristics scores before and after treatment with Betnesol mouthwash QDS/BD for 12 months

Table 14 Summary of Ulcer Severity Scores (USS) before and after treatment with Colchicine 500mcg OD for 12 months in patients with Minor and Major RAS

Table 15 Summary of ulcer characteristics scores before and after treatment with Colchicine 500mcg OD for 12 months

Table 16 Summary of Ulcer Severity Scores (USS) before and after treatment with Colchicine 500mcg OD and Betnesol mouthwash QDS for 12 months in patients with Minor and Major RAS

Table 17 Summary of ulcer characteristics scores before and after treatment with Colchicine 500mcg OD and Betnesol mouthwash QDS for 12 months

AIMS OF THE STUDY

1. The primary aim was to assess the efficacy of Betamethasone mouthwash (topical steroid) and systemic Colchicine, individually and combined, in the treatment of Recurrent Aphthous Stomatitis (RAS).

2. The secondary aim was to assess the effect of these therapies on specific immunological markers (cytokines) to determine the mode of action of each medication.

3. The third aim was to identify the cytokines related to RAS in serum and saliva, during different stages of the disease and to recognise the effect of local steroid and systemic Colchicine treatment on these cytokines.

CHAPTER ONE

LITERATURE REVIEW

RECURRENT APHTHOUS STOMATITIS

1.1 Introduction:

Recurrent Aphthous Stomatitis (RAS) is one of the most common oral mucosal diseases, affecting approximately 5 -10% of the population (Sircus et al 1957). Hippocrates, the father of medicine in 460-370 BC, was the first one to use the term "Aphthae" in relation to a disorder of the mouth. Aphthae is a Greek word meaning ulceration but the first clinical description appears to be used by Mikulicz and Kummel in 1898 (Sircus et al 1957). RAS is defined as recurrent, single or multiple, painful ulcerations in various regions of the oral mucosa without evidence of ulcerations anywhere else in the body or signs of systemic diseases (Tappuni et al 2013). The condition usually arises in childhood or adolescence, and may have a positive family history. These patients may develop oral RAS at an early age and have more severe symptoms than affected individuals with no family history (Ship et al 1972). There is a high correlation of RAS in identical twins, as was shown in a study that included 19 set of twins and 318 individuals from six families (Miller et al 1977). Furthermore, children of high socio-economic status are more commonly affected (19%) than those from low socio-economic groups (2%), as was shown in a study which included 846 Argentinean children aged 4-13 years (Crivelli et al 1988). However, the numbers of boys (555) and girls (291) were not matched in this study; boys were nearly double the number of girls (Crivelli et al 1988).

The occurrence of RAS in the general population ranges between 5-20% (review by Scully et al 2008). However, the prevalence of RAS varies in populations depending on the diagnostic criteria scheme adopted in different research centres and on the patient's ethnic origin (Fahmy 1976). A large cross-sectional study in 1985 found RAS in 2% of Swedish adults of 20333 individuals aged 15 years or more with a history of 2 years of recurrent oral

11 ulcers (Axell et al 1985). A study by Fahmy in 1976, which included 20000 individuals who had lived in Kuwait for over 5 years, showed that 0.5% of Bedouin Arabs were affected by RAS. The study showed that RAS affected 18% of Kuwaitis and 35% of non-Kuwaitis, where 5470 (27%) manifested with recurrent oral ulcerations (Fahmy 1976). In a further study oral examination was performed on 17235 North American subjects (age 17 years and older). Of these, 146 (0.89%) had at least one apparent lesion of RAS. Annual reported prevalence showed that RAS affected 20.8% Caucasians, 12.8% Mexican-Americans and 4.9% Afro-Americans (Rivera-Hidalgo et al 2004). Moreover, other studies showed RAS affected: 1.9% of 308 Spanish adults (Vallejo et al 2002), 1.53% of 8866 Indian subjects (age 15-75 years) (Bhatnagar et al 2013) and 0.5% of Malaysians but only 0.1% of Indians in Malaysia in a study which included 11697 subjects with a mean age of 44.5 years (Zain 2000).

In conclusion, reported prevalence of RAS varies in literatures according to the selection of patients and whether the presence of ulcers at examination or a history of lesions was reported. Many studies were based on non-probability samples, which might result in a significant variation in reported prevalence. (Table 1)

Table 1: Prevalence of RAS

YEAR COUNTRY AUTHOR NO.pats F& M AGE RESULTS COMMENTS

1976 Kuwait Fahmy MS 20 000 6:5 Adults 27% RAS 5% Beduin 18% 22% Urben Kuwai 35% non Kuwaiti

1977 USA Miller MF 1 788 F University Students ↑ Retrospective 57% students study M 48%

2012 Thailand Krisdapong 1100/12y F>M 12-15 24.7%/12y Children S 871/ 15y year 36.2%/15y

2010 Italy Majorana 10128 F=M 0-12 year 14.8% Retrospective A children RAS 1997-2007

1988 Argentina Crivelli 846 M>F 4-13 year 19% S1 S1 high socio MR 463/S1 children 2% S2 S2 low socio 383/S2

1985 Sweden Axell T 20 333 F>M Over 15 17.7% years RAS

1957 United Sircus W 1 738 F>M Adults 19.3% Kingdom and RAS children

One study found that the prevalence of RAS in American subjects younger than 40 years old was 22.5%, while for those older than 40 years it was 13.4% (Hidalgo et al 2004). Another study that included 665 German subjects aged 35-44 years showed that the frequency of RAS during examination was 1.4%, reaching 18.3% when data from patient's history were included (Reichart 2000). In addition, similar results were seen in younger ages in a study which included 1537 high school students aged 13-18 during a school examination, where RAS clinically affected 2% but increased to 27.3% when a history of ulcers were included in the data (Gorska 1997). Nevertheless, the prevalence of RAS in children was higher than adults, which was also demonstrated in a recent study by Krisdapong et al (2011) where 1100 (age 12 years) and 871 (age 15 years) Thai children displayed a prevalence of 24%-36%. These findings were confirmed in a much larger retrospective study that included 10128 Italian children aged 0-12 years old, which showed that 14.8% of the children had a history compatible with RAS (Majorana et al 2010). Since most RAS lesions last for 10-14 days with about 4 weeks before the next batch, the ulcers would be present clinically for only 25% of the time. Thus, the studies that relied on the presence of lesions on examination will always be far less than the prevalence of RAS diagnosed by history.

However, there were extensive limitations in the majority of these studies with a vague definition of RAS, as it was not apparent whether subjects with recurrent oral ulcerations due to systemic diseases or underlying causes were included in their studies which might have affected their conclusions.

Nevertheless, RAS in some literatures could be found as RAU (recurrent aphthous ulcerations) or ROU (recurrent oral ulcerations) or RAS-like disease which is defined as recurrent oral ulcers with local and/or systemic predisposing factors or underlying disorders such as haematinic deficiencies, smoking cessation, hormonal changes, food hypersensitivity, Behcet disease or gastrointestinal disorders (Natah et al 2004). However, the evidence to support these causative factors is debatable (Scully et al 2008). Olson et al 1982, in a study which included 90 RAS patients and 23 healthy controls, reported that there was no significant difference between the groups in serum 14 levels of vitamin B12, Folate and iron since only 3 out of 90 RAS patients proved to have haematinic deficiencies. The authors suggested that haematological tests other than FBC (Full Blood Count) were not indicated for RAS patients (Olson et al 1982). In addition, Lalla et al in 2012 demonstrated in a randomised, placebo-controlled, double-masked, parallel-arm, clinical trial that daily multivitamin supplements did not improve the number or duration of RAS episodes in 160 subjects (Lalla et al 2012). This suggests that there is a small difference between RAS of unknown aetiology and RAS-like disease with predisposing factors since treating the underlying disorder may not result in the amelioration or remission of the ulcers (Challacombe et al 1977, Letsinger et al 2005, Lalla et al in 2012).

1.1.1 Gender and RAS

A slight predominance of RAS has been found for females (Axell et al 1985, Bhatnagar et al 2013). These findings have been confirmed in a 12 year retrospective study which investigated 1788 professional students in Pennsylvania (USA) and showed that 57.2% of the students affected with RAS were female, and 48.3% male. However, only 651 subjects were available for the follow-up (Miller et al 1977). Similar findings were shown in a more recent study by Bell et al in 2005, in which 53% of 1335 RAS subjects were female. In contrast, Hassan et al in 2010 had a ratio of female to male of 1:4 in a clinical trial in Bangladesh, probably due to the limited access of females' patients to the health care facilities. Sircus et al in 1958 reported that there were twice as many women than men in 120 RAS patients. In a retrospective study of complex aphthosis, there were 47% female and 52% male (Liang et al 2012) and Fahmy in 1976 demonstrated a ratio of 6:5 (Female to Male) in 5470 RAS subjects.

In summary, most of the above studies showed that there was no significant predominance of females in RAS patients and there seems to be no evidence that the severity of RAS differs in males and females.

1.1.2 Hormonal changes and RAS

A small number of females suffering from RAS reported cyclical oral ulcers related to the luteal phase of menstrual cycle. This was postulated to be related to changing levels of progesterone and thus defective oral mucosal epithelial turnover (Ferguson et al 1984). This was confirmed recently in a study involving 40 healthy women with RAS and with normal menstrual cycles of 28-30 days. 30% had recurrent oral ulcers during their period (Balan et al 2012). However, this conclusion was contradicted in a retrospective study by McCartan et al 1992, who used MEDLINE database searches from 1988- 1991 to identify articles since 1966 regarding recurrent oral ulcers and menstrual periods, pregnancy, progesterone and oestrogen hormones. This author reported that there was no evidence of an association between RAS and menstrual periods, pregnancy or menopause (McCartan et al 1992). However, these contradictory results from different researchers suggest a need for further investigations with bigger sample size studies.

1.1.3 Stress and RAS

Many studies have reported an association between ulcer recurrences and a variety of psychological factors including anxiety, stress and depression. In a study of relaxation/imagery treatment programmes, a significant decrease in the frequency of ulcer recurrences has been noticed in all treated RAS patients (Andrews et al 1990). This finding has been confirmed in a randomised controlled trial that included 160 American patients classified as mental or physical stressors. Their conclusions suggested that mental stressors were more strongly associated with RAS than physical stressors (Keenan et al 2013). These findings were supported in a study by Huling et al in 2012 which included 160 RAS patients. Their conclusion was that the stressful events were significantly associated with the onset of RAS. However, these findings were based on weekly surveys for up to one year which were done over the phone. Similar results showing an association between ulcer recurrences and stress were demonstrated in Israel (Zadik et al 16

2012) and in Jordan (Al-Omiri et al 2012). Furthermore, Gallo et al in 2009 concluded that psychological stress may trigger RAS rather than being the cause of the disease in a study of 25 RAS patients during their active phase compared with 25 healthy controls. In contrast, other studies have failed to prove any association of anxiety, depression and recurrences of RAS (Pedersen 1989, Ferguson et al 1984).

The determination of anxiety or stress can be detected by salivary cortisol levels which were found above normal in only 3 cases out of 20 RAS patients during active phase compared with 10 healthy controls (Eguia-del Valle et al 2013). Despite the extensive amount of research that has been allocated to the association of RAS and stress, the majority of the studies produced contradictory results and therefore were unable to validate the correlation between them. (Table 2)

Table 2: Stress and RAS

YEAR COUNTRY AUTHOR NO. METHODS RESULTS COMMENTS PATS 2013 USA Keenan 160 RLCQ Yes Stress Randomised AV Questionnaire Mental more controlled than physical stress 2013 Spain Eguia-del 20 Cortisol level Not Valle A RAS (saliva) significant ELISA 10 control

2012 Croatia Picek P 30 STAI Not ROU depression significant 30 inventory test control

2012 USA Huling 160 RLCQ Stress Mental more LB Questionnaire initiate RAS than physical 12 months 2012 Jordan AL-Omiri 50 HAD Yes anxiety F> M MK RAU Questionnaire & RAS 50 control 2012 Israel Zadik Y RAS Yes anxiety CASE pat & RAS CONTROL healthy co 2009 Brazil Gallo 25 Questionnaire Yes CDe B RAS physiological 25 stress & control RAS 2012 India Handa R 50 HAS Yes stress & During & RAU Questionnaire RAS after exams 30 control 2009 Jordan Safadi 684 Questionnaire 50% yes RA stress& RAS

1.2 Clinical Features:

RAS is characterised by the recurrent appearance of round, ovoid or elliptical ulcers often covered by fibro-membranous slough surrounded by an erythematous halo. The lesions are self-resolving with or without scarring after periods of remission (Rogers, 1997). It can be classified as 3 forms: Minor, Major and Herpetiform (Lehner 1968), but it should be noted that these classifications of RAS are clinical diagnoses and not descriptions of individual ulcers (Challacombe et al 1991).

1.2.1 Minor RAS

Minor RAS is a very common type which affects 80% of RAS patients. It is characterised by small oral ulcers 2-4 mm in diameter and their healing period takes no longer than 10-14 days without scarring. The most affected area is non-keratinized mucosa such as labial and buccal mucosa, the floor of the mouth and lateral border of the tongue (Bag ān et al 1991).

1.2.2 Major RAS

Major RAS is the severe form which affects 10-15% of RAS patients. It is characterised by large, oval ulcers more than 10mm in diameter which affected keratinised mucosa like palate, attached gingivea, dorsum of the

19 tongue and tonsils. Healing takes more than 3 weeks and often heals with scarring (Bag ān et al 1991).

1.2.3 Herpetiform RAS

Herpetiform RAS is the less common type which affects 8% of RAS patients. The ulcers are very small (about 2-3 mm in diameter) but multiple in numbers (10-100) and they tend to merge with each other forming large, irregular and painful ulcers. Herpetiform RAS may arise anywhere on the oral mucosa; the healing period takes 10-14 days and may leave scarring if there has been extensive ulceration (Bag ān et al 1991). They were first described by Cooke in 1969 (Cooke 1969).

1.2.4 Other classifications of RAS

Bag ān et al in 1991 studied 93 RAS patients for 6 months without treatment, and classified them into 66 Minor, 20 Major and 7 Herpetiform (Lehner 1968). The study evaluated the number of ulcers and rate of recurrences and then classified RAS into: Type1 (recurrences separated by over 3 months), Type 2 (episodes every 1-3 months) and Type 3 (continuous lesions). Bag ān concluded in this study that patients’ ages were found to be greater for Type 1, followed by Type 3 and then Type 2. As for the duration and number per episode, it was greater in Type 3, followed by Type 2 and then Type 1. Evolution time was longest for Type 1 and shortest in Type 3. Although, Bag ān tried to establish this alternative classification to RAS, the site and location of the ulcers had not been considered which makes it debatable whether it can be complementary to the Minor, Major and Herpetiform classification (Lehner 1968, Bag ān et al 1991).

Another classification of RAS which is less used in the UK is that of Simple and Complex aphthosis (Jorizzo et al 1985). Simple aphthosis is characterised by few lesions with minimal pain that heals in 1-2 weeks and is limited to the oral cavity, while complex aphthosis is the severe form with numerous large ulcers, marked pain and associated with genital lesions. However, complex aphthosis might also represent an atypical form of Behcet Disease and a follow-up of such patients may disclose more complete expression of Behcet Disease in the future. Furthermore, in a follow-up study lasting for 16 months, one patient out of six with Complex aphthosis developed Behcet Disease (Jorizzo et al 1985). These findings were confirmed in 35 Korean patients followed for 7.7 years, in which 52.2% developed Behcet Disease (Bang et al 1995).

Behcet’s Disease was first described by Behcet in 1937 as a chronic, multisystem, vasculitis disorder characterised by oral and genital aphthae, arthritis, cutaneous lesions and ocular, gastrointestinal, and neurological manifestations (Schreiner et al 1987). However, ocular involvement is more frequent in Japan and Middle East while oral and genital ulcers are more 21 common in Europe. The mean age of onset ranges from mid 20s to late 40s, with a higher male to female ratio of Mediterranean descent (International Study Group for Behcet’s Disease 1990). Behcet’s Disease, also known as Silk Road disease, is strongly associated with Major Histocompatibility Complex (MHC) antigen HLA-B51 (review by Verity et al 1999).

In conclusion, classification of Minor, Major and Herpetiform (Lehner 1968) was the superior classification which accommodated all of the ulcer characteristics of size, number, duration and site.

1.3 Genetic factors:

The development of RAS is dependent to an extent on genetic predisposition, although no specific mode of inheritance was established. Previous studies reported that more than 40% of RAS patients may have a positive family history and that these patients may develop oral ulcerations at an early age and have more severe symptoms than affected individuals with no family history (Ship 1966, Sircus et al 1957). Furthermore, Miller et al in 1977 reported that there was a high correlation of RAS in identical twins, in a study that included 19 set of twins and 318 individuals from six families (Miller et al, 1977). In a more recent study of 684 Jordanian RAS patients, family history was reported by 66% (Safadi 2009).

Numerous studies have reported the possible association of RAS and Human Leucocytes Antigen (HLA) sub-types. In a study of Turkish RAS patients, the frequency of HLA-B5 was raised, but not significantly compared with healthy controls (Ozbakir et al 1987). Studies have shown a negative association of HLA-DR4 in Greek RAS patients (Albanidon-Farmaki in 1988) and HLA-B5 in Sicilians, but HLA-DR7 significantly increased in RAS Sicilians (Gallina et al 1985). Although other studies failed to demonstrate an association between HLA haplotype and recurrent aphthous ulceration (Dolby et al 1977), Challacombe et al in 1977 studied HLA haplotype in 120 RAS patients and 65 with Behcet’s disease (BD) compared with 200 controls. Behcet’s disease

22 patients then divided into 4 groups according to the disease manifestations: (1) Muco-cutanous involved oral and genital, with or without skin manifestations (2) Arthritical (3) Neurological (4) Ocular. Further division of Behcet’s disease patients was made according to their nationality, into 52 Caucasoid British and 13 non-British. The study showed a significant association of HLA-A2 and HLA-B12 with ROU (defined as an old term for RAS) while in Behcet’s disease, HLA-B5 was related to the ocular type, HLA- B27 to the arthritic type and HLA-B12 to the muco-cutaneous type of BD. In addition, this study revealed that the frequency of HLA-B5 was higher in non- British (69%) than British patients (19%) but the frequency of HLA-B27 and HLA-B12 was higher in British than non-British (Lehner, Challacombe et al 1979). Moreover, Ohno et al in 1975 reported that 75% of 44 Japanese patients with Behcet’s disease had HLA-B5 but O’Duffy in 1976 could not confirm these findings in 26 American patients.

Furthermore, Lehner et al in 1977 found HLA-B5 in 15% of Behcet’s disease patients and in 12% of healthy controls in a study included 33 British patients. Chang et al in 2001 found that HLA-B51 was present in 56% of 61 Korean patients with Behcet’s disease and 16% in 56 RAS patients compared with 70 healthy controls. Shohat-Zabarski et al in 1992 found that HLA-B51 was present in 23% of RAS subjects and in 9% of healthy controls while HLA-Cw7 was found in 23% of RAS and 5% of controls in a study that included 26 Israeli RAS patients. In a more recent study of 22 Israeli Arabs students found that HLA-B52 was positive in 31.4% while HLA-B44 was positive in 36.4% (Jaber et al 2001).

In conclusion, different studies have reported either the presence or absence of association between HLA and RAS, which could be due to different ethnic backgrounds of the patients. Therefore, further studies are needed to confirm the association of specific HLA haplotype and RAS.

Additional support of a genetic basis of RAS was given by the result of a recent study by Guimaraes et al in 2006, which compared 62 Brazilian RAS patients with 62 healthy controls. The study demonstrated that polymorphisms 23 of high IL-1beta and TNF-alpha production were associated with an increased risk of RAS development. Positive family history was reported in one-third of patients with RAS (reviewed by Jurge et al 2006). This has been supported by another study that included 155 Turkish individuals and found that the frequency of IL-1beta was significantly higher in RAS Turkish patients (Akman et al 2007). However, these findings have been contradicted by Bazrafshani et al in 2002, who showed that there was no significant association between inheritance genotype of TNF-a-308, TNF-b-Ncol and VDR polymorphisms and susceptibility to RAS in 95 RAS subjects (Bazrafshani et al 2002). The same authors in another study, which included 91 RAS patients and 91 healthy controls, demonstrated significant associations of IL-1beta and IL-6 in RAS subjects. Although the sample size was reasonably big in both studies, all of the subjects were Minor type of RAS (Bazrafshani et al 2002).

Kalkan et al in 2013 demonstrated for the first time a significant increase of MEFV gene in a cohort of Turkish patients, which included 100 unrelated patients with a clinical diagnosis of RAS compared with 156 healthy controls (Kalkan et al 2013). Furthermore, Kalkan et al in another study compared 188 RAS patients with 200 healthy controls reported the association of MTHFR gene C677T mutation with the number of oral ulcers in RAS patients (Kalkan et al 2013). In a very recent study by Karakus et al in 2014, which compared 184 Turkish RAS subjects with 150 controls, it was shown that genotypes of IL-6 gene -572G>C and -174G>C polymorphisms were significantly higher in RAS subjects compared with healthy controls. These contradictory findings suggest that confirmatory studies in different populations are needed to determine the role of these genes in RAS patients. (Table 3)

Table 3: Genetics and RAS

YEAR COUNTRY AUTHOR NO.pats GENE RESULTS COMMENTS

2007 Brazil Guimaraes 64 RAS IL-1 b IL-1 b ↑ Mi&Ma AL 64 +3954C/A TNF-a ↑ RAS control TNF-a-308 G/A 2002 Manchester Bazrafshani 95 RAS TNF-a308 TNF-a Mi RAS UK MR TNFB Ncol TNF-b VDR intron VDR 8 & exon 9 Not association 2006 Brazil Guimaraes 62 RAS IL-1 b IL-1 b ↑ Mi&Ma AL 62 (+3954)gene RAS control

2001 Korea Chang HK 61 BD HLA-B51 BD 55.7% Case control 56 RAS RAS 16.15 HLA-B51 70 Cont 15.7% more with control uvietis (BD) 2013 Turkey Kalkan G 100 MEFV gene MEFV ↑ in RAS RAS (Mediterrane 156 an familial control fever) 2013 Turkey Kalkan G 188 MTHFR MTHFR RAS C677T gene C677T gene ↑in 200 RAS control 2002 Manchester Bazrafshani 91 RAS IL-1 a IL-1 b ↑ Mi RAS UK MR 91 IL-1 b IL-6↑ control IL-1 RN IL-1 RN ↑ IL-6 174 2008 Turkey Akman A 155 IL-1a IL-1 a Turkish IL-1 b 889C ↑ Il-1 B 3962T ↑

1.4 Immunopathogenesis:

RAS represents a very common oral mucosal disorder that is poorly understood. However, studies of the lesional biopsies from RAS have shown that in the pre-ulcerative phase, the earliest changes consist of focal degeneration of individual supra-basilar epithelial cells progression to form minor intraepithelial vesicles. Later, but still in the pre-ulcerative stage, a dense lymphocytic and monocytic infiltrate, associated with damaged epithelial cells (Lehner et al 1969, Poulter et al 1989). In this stage Class I and II Major Histocompatibility Complex (MHC) antigens can be found over the epithelial basal cells and more deeply within the epithelium at the ulcer stage, consistent with cell-mediated inflammation (Savage et al 1986).

Although the role for autoimmunity in RAS pathogenesis was first suggested in the 1960s (Lehner 1968), the exact pathogenesis of this mucosal disorder is still unknown. The high load of micro-organisms that colonised the oral mucosa might initiate the immune response, and it was therefore suggested that bacteria like Streptococcus sanguis colonised the oral mucosa and their antigens might be responsible for the initiation of the pathological immune response in RAS. Cross-reactivity between mycobacterial 65-kDa Hsp (Heat Shock Protein) and Streptococcus sanguis was observed and significantly raised serum antibodies to recombinant 65-kDa mycobacterial Hsp have been reported in RAS (Lehner et al 1991). Moreover, Hasan et al in 2002 demonstrated that lymphocytes of RAS subjects during ulcerative stage, compared with the remission period, significantly raised lympho-proliferative response to peptide epitope 91-105 of the 65-kDa mycobacterial Hsp. This study hypothesized that the high load of micro-organisms that colonize the oral mucosa may initiate an immune response by the microbial HSP65 derived peptide 95-105, stimulating the numerous Langerhans cells in the oral mucosa to activate a cross-reacting immune response to the homologous peptide 116-130 within the epithelial HSP60, and thus initiating the immuno- pathological changes that lead to RAS (Hasan et al 1995).

However, other studies have suggested that the role of TLR2 (Toll-like receptor) stimulating PBMCs (peripheral blood mononuclear cells) could be involved in the pathogenesis of RAS (Barros et al 2010). TLRs are a group of membrane receptors which can recognise molecules derived from bacteria, viruses and fungi that are involved in both immune regulation and control of epithelial barrier integrity. The authors suggested that RAS occurred due to an imbalance of Th1/Th2 immune response as well as epithelial barrier dysfunction of the oral mucosa caused by impairment of TLR2 pathways, which permitted the contact of immune competent cells from the lamina propria with oral antigens which are rich in Th1 cytokines and that will influence the onset of RAS (Barros et al 2010).

Despite the two different theories for the pathogenesis of RAS, the exact mechanism of this mucosal disorder is still unknown and further studies need to be done, but with a precise definition of RAS which was not apparent in the previous studies.

1.4.1 Role of cytokines in RAS:

One of the factors involved in the pathogenesis of RAS is possibly a cell- mediated immune response in which several cytokines seem to play a major role (Albanidou-Farmaki et al 2007, Sun et al 2000). Type 1 pro-inflammatory cytokines like interleukin-2 (IL-2), interleukin-12 (IL-12), interferon-gamma (IFN-g) and tumour necrosis factor-alpha (TNF-a) have been suggested to be implicated in the etiopathogenesis of RAS (Lewkowicz et al 2005).

Tumour necrosis factor-alpha (TNF-a) is an important inflammatory mediator and critical cytokine for adequate host defences. Sun et al in 2006 showed that the serum level of TNF-a was significantly raised in a study which included 146 RAS patients compared with 54 healthy controls. The study divided RAS into exacerbation and post exacerbation stages, with further division into Minor, Major and Herpetiform RAS. Although the authors suggested that raised levels of TNF-a could be an indication of the severity 27 and the stage of disease (Sun et al, 2006), it was found to be raised in only 29% (Sun et al 2006). Another study by Natah et al in 2000 supported the previous findings in lesional biopsies of 12 RAS patients compared with 10 controls. However, all RAS subjects were Minor type only which narrowed their findings to one type of RAS. In addition, Buno et al in 1998 demonstrated raised levels of IFN-g, TNF-a, IL-2, but not IL-10, in lesional biopsies of 27 RAS patients compared with 13 controls. The study group divided into 2 parts: part 1 consisted of 21 RAS patients and 7 controls while part 2 consisted of 6 RAS patients and 6 controls. The lesional biopsies were obtained from part 1 within 72 hours of ulcers attacks and from part 2 at 24 and 48 hours after surgical trauma to the oral mucosa. However, there was no matched saliva or serum to support their findings. Contradictory results were shown by Thornhill et al 2007 in a more recent study which included 19 RAS subjects. This study showed that there were no significant differences of TNF-a concentrations in plasma, as TNF-a levels were within the normal in all stages of the disease (Thornhill et al 2007).

Boras et al (2006), in a study which included 26 RAS subjects compared with 26 healthy controls, demonstrated that salivary IL-6 concentrations were not significantly different between acute and remission phases of RAS, while salivary levels of TNF-a were significantly different in the acute stage of RAS compared with controls and between acute and remission phases where the highest concentrations of TNF-a were seen during remission period. This might indicate a possible role of TNF-a in the healing of ulcers (p<0.001). However, interpretation of these results might be affected by the small sample size of 13 patients as 50% of the patients reached the remission period (13 out of 26) (Boras et al 2006). These findings have been recently contradicted by Albanido-Farmaki et al in 2007, who showed that T cells secreting IL-2, IL- 12, IFN-g and IL-10 were increased in 32 RAS patients during the active phase compared with 40 healthy controls (all patients were not on any treatment and their ulcers healed spontaneously), while TNF-a, IL-5, IL-6 were not significantly different and IL-4 levels were decreased. Detailed analysis of this data revealed that the study included Minor RAS only and

ignored Major or Herpetiform, which narrowed their findings to one type of RAS.

Sun et al in 2004 demonstrated raised serum concentrations of IL-6 and IL-8 in a study which included 146 RAS patients during the active stage of the ulcers compared with 54 controls. IL-6 levels were greater than normal in 37 out of 146 (25%) of RAS subjects, of which 19% (13 of 69) of Minor, 33% (20 of 61) of Major and 25% (4 of 16) of Herpetiform RAS. IL-8 concentration was greater than normal in 60% of RAS patients (87 of 146), of which 59% (41 of 69) of Minor, 59% (36 of 61) of Major and 63% (10 of 16) of Herpetiform RAS. This study suggested that serum IL-8 was a more sensitive marker than IL-6 in monitoring the disease activity of RAS. However, the active stage of the ulcer was considered from the first day of the onset to complete healing. (Table 4)

Table 4: Cytokines and RAS

Name of Jou. Date Author Findings Relation to RAS J Oral Pathol 2012 Pekiner FN Same IL-2 30 RAU Med same IL-6 serum Med Oral Patol 2011 Egui-del Inc TNF-A 20 RAS Cir Bucal Valle A serum ImmunolLett. 2008 Curnow Inc. IL-15 20 RAS Slightly Inc IFN- Serum(ELISA) g Slightly Inc IL-2 Same IL-4 Same IL-10 Tohoku J Exp 2007 Albanidou- Inc. IL-2, 12, 10, 32 RAS Med Farmaki IFN-g. 40 control Dec. IL-4 (During active Same. TNF-a, IL- phase of minor 5, IL-6. RAS). Georgian Med. 2007 Koridze Kh Inc. IL-6 61 RAS News Oral path.Med. 2006 Boras Inc. TNF-a 26 RAU (salivary) 26 control Same. IL-6 ( Saliva during (salivary) active and remission phase). Scand J Nov 2006 Dalghous Inc. IL-12, TNF- 19 RAS

Rheumatol a, INF-g 6 control Same. IL-4 (Biopsy). Oral Path. Med. Feb.2006 Sun Inc. IL-6, 8, TNF- 146 RAU a 54 control (Serum in ulcerative stage of minor, major and herpetiform). Immun. Letter Jun.2005 Lewkowicz Inc.IL-2, 5, 8, 6, 10 RAU TNF-a, INF-g. 12 control Dec.IL-10 (Blood during active and remission stage). Am.J.Chin 2005 Sun Inc.IL-2, 6,10, 19 RAU TNF-g, INF-a Oral Path.Med. Mar.2004 Sun Inc. IL-8, IL-6 146 RAU 52 control (Serum, during active stage of minor, major and herpetiform). Oral Path.Med Apr.2003 Sun Inc. IL-6 197 RAU 77 control (Serum). J.Oral Path. 2000 Natah SS Inc. TNF-a 12 RAU 10 control (Biopsy) ProcNatlSci 2000 Sun Inc. IL-2 34 RAS china 32 control (Serum, during different stages of RAS)

ClinExpImmunol 1999 Freysdottir Inc INF-g 11 RAS Same TNF-a 15 control (Serum) Archives of 1998 Buno Inc.IL-2, INF-g, Part 1: 21 RAS Dermatology TNF-a 7 control Same. IL-10 Part 2: 6 RAS 6 control (Biopsy) Asian Pac.J 1998 Bachtiar Inc. CD8+ 19 RAS Allergy Immuno. Dec.CD4+ 8 control Same.CD19+, CD16+/CD56+ Oral Sur oral 1994 Yamamoto Inc.IL-4, IL-6 20 RAS path oral med Same TNF-a 20 control

In conclusion, studies have shown contradictory results on the correlation of different cytokines in serum, saliva and lesional biopsies of RAS, which might be due to the different techniques used in analysing these cytokines giving different results. However, RAS seems to be highly related to the increased level of TNF-a, IL-6, IL-2 and INF-g. In addition, increased levels of IL-10 were demonstrated in two studies only, while in others it was either slightly decreased or remained unchanged. As for IL-8, its level was shown to be highly related to RAS by Sun et al only. IL-12 and IL-5 were slightly related but IL-4 was either decreased or unchanged.

Further studies are needed to determine the specific cytokines related to RAS and whether these cytokines are associated with the presence of the ulcers or with the severity of the disease.

1.4.2 Peripheral Blood Mononuclear Cells (PBMCs):

Lewkowicz et al in 2005 showed an increased production of Type 1 cytokines IL-2, IFN-g and TNF-a as well as IL-5, IL-6 and IL-8 by peripheral blood mononuclear cells in RAS patients during active and remission phases. In contrast, levels of IL-10 and TGF-beta as anti-inflammatory cytokines were decreased in 10 RAS patients compared to 12 healthy individuals. In addition, the author proved that CD4+CD25+ T regulatory cells were decreased in RAS patients and represented 0.654% of CD4+ T cells in the active phase and 0.561% of CD4+ T cells in the remission stage. Although the study suggested that imbalance in pro-inflammatory and anti-inflammatory cytokines may lead to the breakdown of peripheral tolerance in RAS patients, the sample size was insufficient as only 8 patients had reached the remission period. In agreement to this study, Taylor et al in 1992 showed increased production of TNF-a in stimulated and un-stimulated PBMCs during the active phase of RAS compared with healthy controls (Taylor et al 1992). However, in contrast to these findings Freysdottir et al in 1999 demonstrated that the percentage of PBMCs producing TNF-a or IFN-g were not significantly increased in RAS patients (Freysdottir et al 1999). 31

Dalghous et al in 2006 demonstrated the presence of pro-inflammatory cytokines in lesional biopsies from 19 Minor RAS, 25 Behcet’s Disease and 6 healthy volunteers. The study found that both CD4+ and CD8+ T cells were present in the oral ulcers of Behcet’s Disease and RAS patients. The T helper (Th) 1 cytokines; IL-12, IFN-g, and TNF-a and the Th1-associated chemokine receptors CCR5 and CXCR3 were increased in both patient groups as compared to normal controls, indicating the involvement of Th1 immune response in the immuno-pathology of both Behcet’s Disease and RAS. However, the expression of the Th2 cytokine IL-4 was found within the oral lesions of Behcet’s disease but not in RAS patients (Dalghous et al, 2006). The data analysis in this study indicates a simultaneous increase in both pro- inflammatory and anti-inflammatory cytokines in Behcet’s disease and only pro-inflammatory cytokines in RAS. These conclusions have been confirmed in a previous study by Fresdottir et al in 1999 demonstrating the same findings in 11 RAS subjects and 20 patients with Behcet’s Disease compared with 15 healthy controls.

Vascular endothelial growth factor (VEGF) is a multifunctional angiogenic cytokine involved in angiogenesis and wound healing (Schroeder et al 1984) which is found in human saliva, as part of major and minor salivary glands secretion. A study by Brozovic et al in 2002, which included 30 RAS patients who were divided into 3 stages of the ulcer (early, acute and remission stages) compared with 27 healthy controls, showed that the level of salivary VEGF was decreased in early and acute phases of the Major type of RAS and increased to normal levels during healing time. In addition, there was a decreased proportion of T and B lymphocytes, reduced leukocyte migration and ingestion, antibody-dependant cellular cytotoxicity (ADCC) and natural killer cell activity (NK). The authors suggested that VEGF could be involved as a potent angiogenesis agent in the pathogenesis of vasculitis in RAS. However, their findings were based on Major RAS only (Brozovic et al in 2002).

In conclusion, the results of most of these studies supported the role of cell- mediated cytotoxicity in immuno-pathogenesis of RAS. 32

1.4.3 Lymphocyte sub-sets:

Previous studies have shown a markedly increased lymphocytic infiltration in lesional biopsies of RAS patients. Sun et al in 2000 demonstrated a raised CD4+/CD8+ ratio as a result of an increase in CD4+ and normal CD8+ cells in 34 RAS patients during the exacerbation stage compared with 32 healthy controls. As the disease progressed, they observed an increase in the percentage of CD8+ cells and a decrease in the percentage of CD4+ cells which lead to a depressed CD4+/CD8+ ratio in the post-exacerbation stage. The significantly decreased CD4+/CD8+ ratio returns to normal in the remission stage. This study suggested that an infectious agent is involved in the disease process of RAS and the alterations of the T cell subset are secondary to the causative factors which supported the role of cell-mediated cytotoxicity in the immuno-pathogenesis of RAS (Sun et al 2000). This study divided RAS patients into 4 stages: (1) Early stage (6 patients), (2) Late active stage (32 patients) which was subdivided into: (a) exacerbation stage (21 patients) and (b) post-exacerbation stage (11 patients), (3) Early convalescent stage (13 patients) and (4) Late convalescent stage (14 patients). Although, the study sample was sub-divided precisely to different stages of RAS, they have not divided them according to the different types of RAS.

In addition, Bachtiar et al in 1998 performed a study which included 19 RAS subjects (12 Minor and 7 Major type) compared with 8 healthy controls. The study showed that CD4+ was lower in RAS patients compared with controls, CD8+ was higher in Major RAS than Minor, CD4+/CD8+ ratio in Major RAS was lower than Minor, but there was no difference in CD19+ and CD16+/CD56+ between the groups. Their conclusions indicated that RAS was highly associated with abnormal proportions of CD4+ and CD8+ which was dependent on the severity of the disease (Minor and Major), but Herpetiform RAS was not included (Bachtiar et al 1998).

1.4.4 Polymorphonuclear (PMN) and Natural killer (NK) cells:

A study by Sistig et al in 2001 suggested that RAS is characterized by consistent changes in the spontaneous migration and ingestion functions of polymorphonuclear cells (PMN) and depressed natural killer (NK) cells activity during the acute stage of RAS compared with healthy controls, which supports the theory of the natural immunity disorder in RAS patients. 51 RAS patients during the acute stage and remission period were included in this study compared with 47 healthy controls. The study showed that significant lower values of B-lymphocyte (CD19) were detected between the acute and remission stages in comparison to the control group. Low percentages of T- lymphocyte (CD3) and (CD4) were noticed in RAS patients compared with healthy controls, while T-suppressor cells (CD8) were unchanged in all three groups. It was concluded that natural killer (NK) cells activity depressed significantly in RAS patients during the acute stage compared with controls. In contrast, Greenspan et al in 1985 showed that there were no significant differences in natural killer (NK) cells activity during active and remission periods in 10 RAS patients compared with 10 healthy controls. However, the sample size was very small which might affect the interpretations of their results. In addition, Koridze et al in 2007 demonstrated that the quantities of CD3, CD4 and NK cells were diminished, the indices of CD8, CD27 and IgA were within the normal and IgM, IgG, IgE and IL-6 were increased in 61 RAS subjects during the exacerbation stage of the disease (Koridze et al 2007).

1.4.5 Behçet's Disease:

Recurrent oral ulceration is one of the main symptoms of Behcet's disease which is a chronic multi-system disorder characterized by oral and genital ulcers, cutaneous, ocular, arthritic, vascular, and central nervous system involvements (Direskeneli 2001). It is a disorder of young adults found predominantly between the Mediterranean and the Orient. Behcet's disease is strongly associated with the Major Histocompatibility Complex (MHC) antigen HLA-B51 (Verity et al 1999) and the association of HLA-B5 in Turkey, Israel

34 and France but not in Britain or the United States (Lehner et al 1977, O’Duffy et al 1976). A study Challacombe et al in 1979, which included 65 patients with Behçet's disease compared with 200 healthy controls, demonstrated the association of HLA-B5 to the ocular type, HLA-B27 to the arthritic type and HLA-B12 to the muco-cutanous type (Challacombe et al 1979).

Although the pathogenesis of Behcet's disease is unknown, there is evidence of genetic, infectious and immunological factors associations (Zierhut et al 2003, Fortune et al 2006). Therefore, cytokines and chemokines might play an important role in the immune response of Behcet's disease (Mege et al 1993, Fortune et al 2006).

Oztas et al in 2005 showed that serum concentrations of IL-18 and TNF-a in 27 patients with active Behcet’s disease were significantly increased compared with 20 healthy controls. Yalçindag et al in 2008 studied 22 patients with Behcet's disease and 19 healthy controls; found that IL-6 levels were significantly higher in Behcet's disease than controls. However, these findings were contradicted in another study by Sayinalp et al in 1996 which measured serum concentrations of IL-2, IL-6 and TNF-a in 24 patients with active BD compared with 15 healthy controls using the ELISA technique. The study showed that TNF-a levels were significantly higher in patients with Behcet's disease but there was no change in serum levels of IL-2 and IL-6 between the groups (Sayinalp et al 1996). Furthermore, Sugi-Ikai et al in 1998 demonstrated that the concentrations of IL-2 and IFN-g in the active stage of Behcet’s disease were significantly higher than those during the inactive phase of the disease. The study included 52 patients with Behcet’s Disease, where 32 patients during the active phase and 20 patients during the inactive stage of Behcet’s disease were compared with 33 healthy controls (Sugi-Ikai et al 1998). In agreement to these findings, more recent studies by Misumi et al in 2003 and Akadeniz et al in 2004 showed that levels of IL-12, IFN-g and TNF-α were significantly increased in patients with active Behcet’s disease (Misumi et al 2003, Akadeniz et al 2004).

Dalghous et al in 2006 demonstrated the increased expression of pro- inflammatory and anti-inflammatory cytokines in 25 patients with Behcet’s disease while only pro-inflammatory cytokines were observed in 19 RAS patients, since significant raised levels of INF-g and TNF-a were seen in lesional biopsies of RAS while the IL-4 level was raised in Behcet’s disease and not in RAS. IL-10 mean levels were increased in both Behcet’s disease and RAS patients but it was not statistically significant compared with 6 healthy controls. However, the healthy controls’ sample size was very small which may have affected their conclusions (Dalghous et al 2006).

1.5 Methods of assessing Ulcer Severity:

Methods for assessing ulcer severity varies between literatures partly due to each research group designing their own method to suit their particular objectives. Researchers have used patient's diaries, clinical photographs, history and clinical examination to monitor the response to treatment. This is due to the lack of widely accepted measures of ulcer severity which are easy to use, can be applied to every RAS patient and most importantly can accommodate all ulcer characteristics like size, number, duration, site and pain.

1.5.1 Methods used in research studies are detailed below:

Sun et al (1994) in their clinical trial of Levamisole in the management of RAS, which included 50 patients, had relied on frequent patients’ history. The study paid particular attention to the frequency, duration, number and pain of the ulcers per month. Taylor et al (1993), in a six week clinical trial enrolled 35 RAS patients who were examined on days 1,3,5,7 and 10 with daily diaries. Hunter et al 1987 recorded the site of each ulcer on a diagram of the oral cavity in the daily diaries. Pain was recorded by 10cm visual analogue scale (VAS) ranging from no discomfort to severe pain. Miller et al (1980) enrolled

50 RAS patients in a double blind clinical trial to test the effectiveness of Carbamide peroxide on healing time and pain relief. The sizes of the ulcers were measured to the nearest millimetre in diameter using calibrated periodontal probes. The degree of pain was recorded as none, mild, moderate or severe. Merchant et al (1978) in a double blind clinical trial used photographed lesions before treatment with Betamethasone 17 benzoate. Patients were seen every two days and re-photographed. The study recorded size, location and appearance of the lesions.

Unfortunately, these methods are not considered practical for universal and routine clinical use. In conclusion, there was no uniform or standard method of assessing the severity of the oral ulcerations.

Nevertheless, recently Tappuni et al (2008), in a study which included 45 patients on Colchicine 500mcg a day and Betamethasone mouthwash for 3 months compared with 45 patients on Betamethasone mouthwash as monotherapy, showed that combined therapy of Colchicine and Betnesol mouthwash significantly improved the ulcer free period of RAS and reduced the number of ulcers, size, duration, site affected and pain score, according to a scoring system that converts RAS symptoms to a numerical values to give each ulcer’s characteristics a score (Ulcer Severity Score, USS) (Tappuni et al 2008). This scoring system aided in monitoring the progress of the condition, helped in assessing the efficacy of any treatment and aid in the management of the ulcers. A more recent study by the same authors included 247 RAS patients (excluding 24 atypical ulcers and ulcers with underlying systemic diseases). Thus, 223 RAS subjects were included (136 Minor, 72 Major and 15 Herpetiform RAS). The USS was developed with the premise that more than 95% of RAS episodes are comprised of ulcers that are: less than 20 in number, less than 20mm in diameter, last for less than five weeks and recur in less than 10 weeks. However, diaries and clinical examination were still used to validate the history given by the patients (Tappuni et al 2013).

Further studies measuring the effect of RAS on quality of life have been carried out by Hapa et al in 2011. This study focused on Oral Health Related Quality of Life (OHR-QoL) of 128 RAS patients compared with 40 controls. They showed that OHR-QoL provided an additional dimension which may help to improve the impact of RAS on an individual's life. OHIP-14 measured the degree of RAS impact on functional limitation, pain, psychological discomfort, physical disability, psychological disability, social disability and handicap. They were scored as: never= 0, hardly ever= 1, occasionally= 2, fairly often= 3 and very often= 4. Pain was evaluated by 5 points of mild, disturbing, moderate, severe and distressing. However, only 3% of the enrolled participants were available for their review and completed the OHR- QoL score after treatment with either Colchicine 1.5mg/day or topical steroids (Hapa et al 2010).

A study by Mumcu et al in 2009 designed a standardised Composite Index (CI) to assess ulcer activity in Behcet’s Disease and RAS. In this cross sectional study, 121 subjects with Behcet’s Disease and 45 RAS patients were enrolled who were in an active stage in the previous 3 months of the trial. Composite Index (CI) included the presence of oral ulcers, pain and functional status in both active and inactive periods of the disease. Although the study presented Composite Index (CI) as a reliable and suitable tool for evaluating the clinical impact in Behcet’s Disease and RAS patients, the index in reality was more applied to Behcet’s disease than RAS (Mumcu et al 2009). However, this Composite Index does not include size or duration of the ulcers. Furthermore, in a recent study by Baccaglini et al in 2013, a new standardised diagnostic process (RASDX) was developed for the purpose of identifying RAS patients on the basis of history and RAS photographs recognition only. RASDX consisted of an initial phone screening using standardized questionnaires and recognition of RAS photographs. However, their initial inclusion was 338 subjects but 54 of them were RAS patients, which raises the question of how reliable the questionnaires over the phone were! (Baccaglini et al 2013).

In conclusion, the Ulcer Severity Score (USS) has shown to be the most reliable method of measuring the severity of RAS since it accommodates a wider range of ulcer characteristics to assess the improvement in the disease and determine the effect of therapy on specific ulcer characteristics, as well as being easy to use. Additionally, the USS reflects the disease severity and monitors any clinical changes with treatment since insignificant changes of the USS are frequently an implication of a lack of response to treatment and an indication to revise the treatment plan and/or diagnosis. It is the most standardised method for the assessment of RAS which includes most of the ulcer characteristics (size, number, duration, pain, site and ulcer free periods). As such, USS with a structured management and specific follow up periods may improve the treatment outcome of RAS.

1.6 Management of RAS:

To date there are no internationally accepted guidelines for RAS treatment despite RAS being one of the most common oral disorders. There have been many attempts over the years to find an effective treatment for RAS, but because of its unknown definite aetiology most forms of treatment consist of therapeutic measures to suppress its symptoms rather than a definitive cure. Nevertheless, choice of treatment depends on the severity of the disease. Although the symptoms can be reduced by topical local measures for severe and constantly recurring ulcerations and immuno-modulating therapies might be used to reduce recurrences, it is difficult to prevent RAS (Ruah et al in 1988, Tappuni et al in 2008).

Managements of RAS can be divided into topical treatments which are designed to help healing, and systemic medications aiming to reduce recurrences.

1.6.1 Topical therapies:

A wide range of topical therapies have been used in the treatment of RAS but there are very few published randomised clinical trials to support their efficacy. In a double-blind study by Matthews et al in 1987 of 18 Minor RAS patients, Chlohexidine gluconate was used as a 0.2% mouthwash. The participants were given either Chlorhexidine, Benzydamine or a placebo mouthwash for a 3 months period in a random order. Ulcer diaries were marked at the same time each week during the 9 month test period and records were made of the number, size, sites and pain severity of each ulcer. The study concluded that there were no significant differences between any of the treatments tested, although 8 patients stated a personal preference for Benzydamine hydrochloride mouthwash because of the transient local anaesthetic effect, which provided transient relief of pain (Matthews et al 1987). However, contradictory findings were shown in a study by Edres et al in 1997 where Chlohexidine gluconate mouthwash was found to be more effective than Benzydamine hydrochloride in 38 RAS patients (Edres et al 1997).

Another study investigated the treatment of RAS with Amlexonax which is an anti-inflammatory, anti-allergic and immunomodulator. A randomised, double- blind, multicentre clinical trial was conducted in a Chinese group (104 Minor RAS patients) to validate the efficacy of Amlexonax oral adhesive tablets applied 4 times a day for 5 days (Liu et al 2006). Their study showed a reduction in size and pain with no side effects reported. However this might be due to the short period of only 5 days and a longer trial period may have different conclusions. Furthermore, Meng et al in 2009 used the same 104 patients from the previous study on Amlexonax oral adhesive tablets compared with 210 patients who used Amlexonax oral adhesive pellicles. There were no significant differences reported between the 2 forms of the treatments in terms of effectiveness but patients preferred the pellicles for the ease of use (Meng et al 2009).

Recently a new approach to the treatment of RAS in-vitro with bio-adhesive gel containing Cyclosporine A solid lipid nano-particles on 36 rabbits has been done by Karavana et al in 2012. The stability of the prepared formula was assessed to be used on buccal mucosa which showed that 64.8% of the formula remained for 6 hours after application. Although this study has shown that Cyclosporine bio-adhesive gel is a promising topical treatment of RAS, proper clinical trials are needed to confirm its effectiveness in-vivo.

Topical Tetracycline mouthwash has been used alone or in combination with Amphotericin with or without topical steroids, especially in the treatment of Herpetiform RAS. Topical Tetracycline mouthwash might reduce the severity of the ulcerations but does not prevent recurrences. These findings have been proved by Graykowski et al (1978) in a double blind study which included 25 RAS patients. Although both placebo and Tetracycline groups experienced reductions in ulcer incidence during the treatment period, the Tetracycline group noticed significant reductions in the duration of ulcers, size and pain. Side effects recorded in patients on Tetracycline were comparable to those on the placebo (Graykowski et al 1978). These findings were supported in other studies by Ylikontiola et al in 1997 and Henricsson et al in 1985. In a more recent clinical, randomised, cross-over study by Gorsky et al in 2007, seventeen patients with high frequencies of RAS were randomly given either topical therapy of 0.2% Minocycline mouthwash or 0.25% Tetracycline mouthwash. The study suggested that Minocycline mouthwash reduced the severity of pain on the visual analogue scale (VAS) significantly more than Tetracycline mouthwash (Gorsky et al 2007). This finding was confirmed later by the same authors in 2008, in a placebo-controlled study which included 33 RAS patients (Gorsky et al 2008).

Furthermore, a randomised, double blinded, multi-centre clinical trial with a large sample size was performed by Zhou et al in 2010, which included 258 Minor RAS patients classified into 3 groups, (86 on Penicillin G potassium, 88 on placebo and 84 on no-treatment controls). All subjects applied the troches 4 times a day for 4 days and were asked to record any improvements in size and pain on days 0, 3, 4, 5 and 6. Although Penicillin G was significantly 41 effective in reduction of pain (p<0.0001) and size of the ulcers (p<0.0001), they narrowed their findings to Minor RAS only and ignored other types of RAS which might affect their conclusions (Zhou et al 2010).

Other topical herbal treatments as alternative therapies have been used such as Aloe Vera gel (Babaee et al 2012), Berberine gelatine (Jiang et al 2012), Yunnan baiyao (Liu et al 2012), Myrtus communis (Babaee et al 2010) and Citrus oil with magnesium salts (Shemer et al 2008). However, all of these topical herbal therapies have been used for the treatment of Minor RAS only and further studies are needed to support their effectiveness in severe forms of RAS.

Topical corticosteroids remain the mainstay of RAS treatment in order to reduce an ulcer's symptoms. Miles et al in 1993 compared Triamcinolone acetonide with Chlorhexidine gluconate in 30 RAS patients followed for 12 weeks. All patients used a diary to record their pain level on a 10cm visual analogue scale (VAS). However, there was no significant difference between the groups in terms of pain and duration of ulcers (p<0.4) (Miles et al 1993). In a more recent study by Liu et al in 2012, a double-blind, randomized, placebo- controlled, parallel, multicentre clinical trial was conducted to compare the efficacy and safety of Dexamethasone ointment in 120 RAS patients compared with 120 controls. Patients were instructed to apply the given ointment 3 times a day for 5 days. The size of the ulcers, pain, healing ratio, duration and adverse events were recorded. The Dexamethasone group showed a significant reduction in size (p<0.001) and pain (p<0.001) alongside an 83.3% improvement in healing time, although an improvement of 54.7% was reported in the placebo group (Liu et al 2012). Their findings were supported by Al-Na’mah et al in 2009 in a study which included 53 RAS patients using Dexamucobase compared with 37 patients treated with triamcinolone acetonide in Orabase. The study showed that there was a significant difference between both groups in healing time (p<0.001) while no difference was reported in the pain score (Al-Na’mah et al 2009). (Table 5)

Table 5: Topical treatment of RAS YEAR COUNTR AUTHOR TREATMENT NO.pat RESULTS COMMENTS Y 2009 Iraq Al- Dexamucobase 53 Duration ↓ Case control Na'mah Triamcenolone 37 Pain (same) ZM acetonide 2012 China Liu C Dexamethasone 120 Size ↓ Randomised ointment Pain ↓ Double blind placebo 120 duration ↓ parallel 2009 China Meng W Amlexanox 216 Size ↓ Randomised pellicles pain ↓ Double blind Placebo parallel

2006 China Liu J Amexanox adhesive 104 Pain ↓ Randomised tablet Size ↓ Double blind Vehicle controlled 108 Erythema(same) parallel efficacy ↓ 2008 Israel Shemer A Citrus oil & 26 duration ↓ Randomised magnesium salts pain ↓ adhesive Benzocaine 22 solution 2007 Israel Gorsky M Minocycline rinse 17 Pain ↓ Randomised Tetracycline rinse Cross over 2008 Israel Gorsky M Minocycline rinse 33 Pain (VAS) ↓ Randomised placebo Cross over 2010 China Zhou Y Penicillin G troches 86 size ↓ Randomised Placebo 88 pain ↓ Double blind No treatment 84 parallel 2010 Iran Babaee N Myrtuscommunis 45 Size ↓ Randomised paste pain ↓ Double blind placebo erythema ↓ controlled

2010 China Liu X Yunnan baiyao 227 Size ↓ Randomised toothpaste pain ↓ Double blind placebo controlled 2012 China Jiang XW Berberine gelatine 84 pain ↓ Randomised placebo size ↓ Double blind erythema ↓ controlled

2012 Turkey Karavana Cyclosporine A 36 Healing time ↓ Vivo SY buccal adhesive experiment (Rabbits) 2012 Iran Babaee N Aleovera gel 40 Pain ↓ Double blind placebo Size ↓ Case control duration ↓ 2012 Iran Fani MM Phenytoin syrup 30 53.3% improve ROU (BD) Triamcinolone acetonide ointment 86.7% improve 1993 India Miles DA Triamcinolone 30 Pain (same) acetonide No. (same) Chlorhexidinedigluc onate control

In conclusion, a wide range of topical corticosteroids therapies have been used for the treatment and management of RAS but there was a lack of published clinical evidence to support their efficacy partly due to the lack of using a standardised uniform method for evaluating RAS before and after treatment. In addition, most of the trials (even with a big sample size) were applied on the Minor type of RAS which might affect the interpretation of their conclusions when severe forms of RAS were included.

1.6.2 Betamethasone sodium phosphate (Betnesol mouthwash):

Betnesol mouthwash is one of the most common treatments used in specialised clinics. It is a recognised therapy for RAS and generally accepted as effective in controlling this common oral condition, despite the limited clinical evidence to support its efficacy.

Betamethasone sodium phosphate is very soluble in water, and therefore it can be used as a mouthwash. It does not normally cause retention of salt and water and the risk of inducing oedema is almost negligible. While particular care is required when considering using it during pregnancy as it might cause a transient suppression of the foetal heart rate and increasing the incident of congenital abnormalities like cleft lip and palate, there is no convincing evidence to support this. During lactation significant levels of corticosteroids may be detected in the breast milk, which could cause a degree of neonatal adrenal suppression (British National Formulary BNF).

Betamethasone sodium phosphate is a potent glucocorticoid steroid. Particular care is required when considering using it systemically in patients with a history of osteoporosis, hypertension, congestive heart failure, diabetes mellitus, liver failure, renal impairment and peptic ulceration. High doses for a long period might cause Cushing's syndrome, which is reversible on withdrawal of the treatment (British National Formulary BNF).

1.6.2.a Betamethasone and RAS:

Betnesol mouthwash has been used for symptomatic relief and treatment of RAS (reviewed by Scully 2005). Betnesol mouthwash is a Betamethasone sodium phosphate tablet 500mcg dissolved in 10 ml of water and used as a mouthwash for 3 minutes then discarded. It is administered four times a day (QDS) in the presence of ulcers and twice a day (BD) in between ulcer attacks (Challacombe and Shirlaw 1991). A three-month study by Tappuni et al in 2008 compared 45 RAS patients using Betnesol mouthwash with 45 RAS subjects using Betnesol mouthwash and Colchicine tablets 0.5mg a day. The authors showed that five ulcer parameters (size, number, duration, pain and site) were significantly decreased in the Betnesol group, while patients on the combined treatment of Colchicine plus Betnesol reported significant improvements in number, size, duration and ulcer free periods (Tappuni et al, Challacombe 2008). These findings have been established in a more recent study by the same authors. A three-month Betnesol mouthwash therapy in 79 RAS patients significantly decreased all of the ulcer parameters except the ulcer free period (Tappuni, Shirlaw, Challacombe 2013). Although continuous use of topical steroids could induce candidal infection, prophylactic anti-fungal treatment might be considered for management (reviewed by Rees et al 1996). However, Hegarty et al (2002), in a cross-over randomised study which included 44 patients with Lichen Planus, used Betnesol mouthwash QDS or Fluticasone spray for 6 weeks with a wash out period of 2 weeks. Only 1 patient out of 44 experienced oral candidosis in the absence of anti- fungal therapy (Hagerty et al 2002).

In conclusion, there were very few published, randomised, controlled clinical trials to evaluate the efficacy of Betnesol mouthwash for the management of RAS; proper clinical trials with a good sample size and a large timeframe need to be done.

1.6.2.b Betamethasone and cytokines:

Betamethasone in various forms and its effect on pro-inflammatory and anti- inflammatory cytokines were suggested in numerous studies. Takahashi et al in 1996 recruited 77 patients complaining from sciatical pain following lumbar disc herniation. The study showed that IL-1 alpha, IL-1 beta, IL-6 and TNF-a were significantly decreased with the use of Betamethasone injection. However, there were no placebos or healthy controls to validate their findings. In contrast, Barton et al in 1991 compared the effect of Momethasone, Hydrocorisone, Betamethasone, Dexamethasone and Beclomethasone on IL- 1, IL-6 and TNF-a in supernatant fluid for the treatment of different inflammatory disorders in-vitro. The study reported that the inhibition of IL-1, IL-6, and TNF-a production by an extremely low concentration of Mometasone furoate, when compared with a high concentration of Betamethasone, suggested that Mometasone is more effective than Betamethasone for the treatment of various disorders according to their inhibition effects on these pro-inflammatory mediators (Barton et al 1991). However, further studies are needed to confirm its effectiveness in-vivo. Irakam et al in 2002 suggested that a combination dose of systemic Dexamethasone and Betamethasone for the treatment of chronic lung diseases in children, and their inhibition effects on IL-8, was more significant than Dexamethasone or Betamethasone alone (Irakam et al 2002).

Furthermore, topical application of steroid ointments like Betamethasone valerate, Prednicarbate and Clobetasol propionate on a full thickness skin model had a slight effect on the induction of IL-6 and IL-8 after application for 14 days, which caused a reduction in number of the epidermal cell layers (Zöller et al 2008).

Inhaled therapy of different types of steroids for the treatment of allergic airway diseases and the effect of these inhalers on different cytokines was performed in a study by Umland et al in 1997. It showed that Mometasone and Fluticasone were more potent than Beclomethasone, Triamcinolone, Budesonide, Betamethasone and that their inhibition of IL-4 and IL-5 was up 46 to 50%, while their effect on IFN-g was not significant (Umland et al 1997). In contrast, Baroody et al in 1998 performed a double blind, placebo-controlled study on 15 patients with seasonal allergic rhinitis. They were prescribed Beclomethasone dipropionate twice a day delivered to one nostril and a placebo to the other for 1 week. Nasal scrapings and biopsies were obtained. The study postulated that treatment with topical steroids inhibited the clinical response but had no effect on the inflammation in the deeper compartments (Baroody et al 1998). However the study needed a larger sample to confirm their findings.

Betamethasone injections were administered to animals in 2006 to monitor the inhibitory effect of steroids on pro-inflammatory and anti-inflammatory cytokines. They found that Betamethasone reduced the elevation of TNF-a and IL-1(beta), and induced the expression of IL-10 in the brain, all of which correlated with the changes of pain thresholds in rats. Their findings suggested that the effects of epidural Betamethasone treatment are due to the inhibitory abilities of pro-inflammatory cytokines and stimulatory effects on anti-inflammatory cytokines in rats’ brains, which inhibits the development of neuropathic pain (Xie et al 2006). Corbel et al in 2002 demonstrated that when mice were treated with Betamethasone 5mg/kg, TNF-α levels were reduced, and the levels of IL-10 were increased. However, Sakuma et al in 2001 studied Tacrolimus ointment in comparison to Betamethasone valerate for the treatment of atopic dermatitis in animal models. The suppressive effect of Tacrolimus on the production of the cytokines involved (IL-2, IL-3, IL-4, IL-5 and IFN-g) was more significant than Betamethasone valerate in serum.

In conclusion, there are very few published studies that evaluate the efficacy of Betamethasone mouthwash for the treatment of RAS. Monitoring the effect of this potent topical steroid on pro-inflammatory and anti-inflammatory cytokines in serum and saliva will improve the understanding of the mode of action of this potent therapy.

1.6.3 Systemic Therapies:

The main goals of systemic therapies are to minimise the duration of ulcers and to reduce the frequency of recurrences. Systemic immunomodulatory medications have therefore been tried for the treatment and management of severe and constantly recurring RAS, such as systemic Prednisolone (Pakfetrat et al 2010), Dapsone (Handfield-Jones et al 1985), Levamisole (Sun et al 1994), Azathioprine (Brown et al 1990), Pentoxifylline (Thornhill et al 2007), Colchicine (Ruah et al 1988) and Thalidomide (Gunzler et al 1992) These may produce remission or reduction in symptoms but all have several side effects. Therefore the treatment choices should be guided by the severity of the RAS and the potential adverse effect of the medications. (Table 6)

Table 6: Systemic treatment of RAS

YEAR COUNTRY AUTHOR TREATMENT NO.PATS RESULTS COMMENTS

2010 Japan Yasui Ascorbate 16 Frequency Cross over (Vit. C) Children 50% ↓ 2000mg/day pain ↓ 2009 Israel Volkov I 1000mg/day 31 Duration ↓ Randomised Vit. B12 Pain ↓ Double blind Control No. ↓ Controlled 27 6 months 2012 UAS Lalla Multivit tab 83 No. Randomised RV Placebo 77 Duration Parallel Pain Double blind 12 months No diff controlled 2010 Italy Femiano Prednisolone 20 No. ↓ Double blind F 25mg/day Pain ↓ 6 months Montelukast 20 duration ↓ 10mg/day Placebo 20 100mg/day 2009 Sao Paulo Weckx Levamisole 14 No. Double blind LL 150mg Duration 6 months tds/week Pain Placebo 10 50% improv (levamisole) 70% improv (placebo) 2010 France Hello M Thalidomide 76 Complete Retrospectiv 3 months remission study 84% side eff 85% 2003-2008

2010 France Gil H Thalidomide 41 Remission Retrospectiv 25mg/day study No side effec 2013 Denmark Sand FL TNF-a 18 Complete 3-77 months inhibitor remission (Infliximab 89% ,adalimumab, golimumab) 28% side eff 2009 Brazil De Clofazimine 23 Remission Randomised Abreu 100mg/day (Clofazimin Partially MA Colchicine 23 e blinded 0.5mg tds 44%) Controlled Placebo 20 Colchicine 6 months Side eff45% 2007 Sheffield Thornhil Pentoxifyllin 26 pain ↓ Randomised (UK) l MH 400mg tds size ↓ Double blind placebo no. ↓ Placebo cont ulcer free ↑

1.6.4 Levamisole:

Levamisole is an anti-helminthic drug which has been tried with promising results in patients with severe RAS, providing long term benefits (Sharma et al 2014 review). It is an immuno-modulator which acts as an immunosuppressant with a prolonged dosage and a potentiator of the immune response at lower or intermittent doses. A study by Sun et al in 1994 demonstrated improvement in pain, duration and number of ulcers in 31 RAS patients on 100-150mg a day for 3 months. Although no side effects were reported in the study, adverse effects of Levamisole are nausea, diarrhoea and agranulocytosis. Lehner et al (1976) in a double-blind, cross-over trial, which included 17 RAS subjects, reported a significant decrease in number and duration of the ulcers after 2 months of treatment with Levamisole. While 64% of patients responded to the treatment, 23% reported an increase in number of their ulcers. Recently, Picciani et al in 2010 used Levamisole 150mg once a week for one month in a case report (with 15 years of history with multiple Major RAS). Although they achieved 6 months ulcer free during the follow up period, the need for intra-lesional steroids injections was mandatory to speed the regression of the ulcer. However, Weckx et al in 2009 showed no significant difference between Levamisole and placebo groups, in a study that included 14 RAS patients used Levamisole 150mg three times a week compared with 10 subjects who were placebo controls. Surprisingly, the study demonstrated an improvement in number, duration of ulcers and pain in 70% of the placebo group and only 50% of the Levamisole group (Weckx et al 2009). Furthermore, Sun et al in 2006, 2004 and 2003 demonstrated the effect of Levamisole on different serum cytokines in RAS subjects (IL-6, IL-8 and TNF-a) during the ulcerative and remission stages of Minor, Major and Herpetiform RAS. The study suggested that treatment with Levamisole for 0.5-3.5 months could significantly reduce serum concentrations of IL-6, IL-8 and TNF-a in RAS patients (Sun et al 2006, 2004, 2003).

1.6.5 Dapsone:

Dapsone has been reported to reduce ulcer recurrences in complex aphthosis with a dose of 100mg/day by Sharquire et al in 1984. The exact mechanism is thought to be due to its anti-neutrophilic mechanism. A recent retrospective study (between 1998 and 2007) by Lynde et al in 2009 showed that Dapsone is an effective and safe therapy for the treatment of complex aphthosis in 55 RAS patients (Lynde et al 2009). However, 5 patients only started receiving Dapsone as monotherapy at 25mg/day for 3 days, 50mg/day, 75mg/day, 100mg/day for one week then 125mg/day, whereas 14 patients were on combined therapy of Dapsone and Colchicine. This study evaluated size, number of ulcers, location and healing time, with an overall outcome measured as moderate response (>50% improvement) or substantial response (>75% improvement). Furthermore, 3 subjects out of 5 who received Dapsone as monotherapy had a 75% improvement while 37% reported side effects of heamolytic anaemia, leukopenia and parasthesia of the face. However, the study was limited in its observational and retrospective design (Lynde et al 2009).

Mimura et al (2009), in an open 4 years clinical trial which included 9 patients with severe types of RAS, started with 25mg/day of Dapsone which was increased by 25mg every 3 days until 100mg a day as a maintenance dose. The efficacy was classified as excellent (no relapses), moderate (relapses but with reduction in number, frequency and pain of the ulcers), mild (reduction of pain only) and no response. Although 89% showed excellent to moderate improvements in their symptoms and 5 reported complete remission during the treatment period, Dapsone had to be discontinued in 6 patients due to adverse events of anaemia, haemolysis and jaundice (Mimura et al 2009). However, their sample size was very small which might affect the interpretation of their results. Nevertheless, contradictory findings were shown in a double-blind, placebo-controlled study by Sharquie et al in 2008 where 45 RAS participants were randomly placed into 3 groups to receive either Dapsone 50mg BD, Zinc sulfate 150mg BD or a placebo. The study revealed

51 that the Zinc sulfate group showed a higher reduction of ulcer diameter than the Dapsone group (Sharquie et al 2008).

In addition, Letsinger et al (2005) included 27 patients with complex aphthosis treated with Dapsone dosed from 25mg/day to 100 mg/day, but 6 subjects had significant anaemia as a side effect of Dapsone which therefore required adjustment of their doses. This study reported that only 2 out of 27 (7%) achieved 50% improvement whereas in the combination of Dapsone and Colchicine group, 12 out of 21 (57%) achieved more than 50% improvement (Letsinger et al 2005). However, managing the signs and symptoms of complex aphthosis is a major concern as these patients might progress to Behcet’s disease, and as such close clinical follow ups with thorough investigations are mandatory.

1.6.6 Azathioprine:

Azathioprine is an immunosuppressive drug which belongs to the chemical class of purine analogues and is used in organ transplant and treatment of autoimmune diseases. Azathioprine was synthesized originally as a cancer drug and as a prodrug for mercaptopurine in 1957. The main adverse effect of Azathioprine is bone marrow suppression which can be life-threatening, especially in patients with a genetic deficiency of the enzyme Thiopurine S Methyltransferase (TPMT). Additionally, Azathioprine can cause nausea, vomiting, diarrhoea, fatigue, thrombocytopenia and leukopenia; it therefore requires regular monitoring of the blood count during treatment periods. Although, epidemiological studies by the International Agency for Research on Cancer (IARC) have provided adequate evidences of Azathioprine carcinogenicity in humans, the risks involved seem to be related to the dosage and the duration of the treatment.

There were very few published studies using Azathioprine for the management of RAS but due to its immunosuppressant effects, it has been used more in the treatment of Behcet’s disease. Brown et al in 1990 reported 52 the effectiveness of Azathioprine 50mg two times a day combined with topical Dexamethasone mouthwash in the treatment of Major RAS in a 32 years old female with a 3 month history of Major RAS. Although the lesion was resolved in 90 days with no adverse effect reported, the study was a case report and further studies are needed to validate their findings (Brown et al 1990). Furthermore, in a double-blind, randomised, placebo controlled study of 73 subjects with Behcet’s disease, Azathioprine has been found to be an effective therapy in the management of oral and genital ulcerations, uveitis and arthritis. In addition, Azathioprine was significantly superior to the placebo group in preventing the development of uveitis. Therefore, the authors recommended that using Azathioprine as a prophylactic therapy to prevent eye involvement in Behcet’s disease (Yazici et al 1990).

However, in a more recent and larger sample size study by Saadoun et al in 2010, it was reported that Azathioprine was an effective therapy in 157 patients with severe uveitis of Behcet’s disease. Following Azathioprine therapy of 2.5mg/kg, 81 subjects (51.6%) were complete responders, 65 (41.4%) were partial responders and 11 patients (7%) were non-responders. Although the authors revealed that Azathioprine was an effective and safe therapy, two patients had hepatotoxicity and one patient had bacterial septicaemia (Saadoun et al 2010).

1.6.7 Thalidomide:

Thalidomide is an anti-TNF-a therapy which was first marketed as a sedative medication and which became an over-the-counter drug in Germany on the 1st of October 1957. Afterwards, Thalidomide was used for the treatment of nausea and morning sickness in pregnant women but soon became infamous for its teratogenicity, therefore its negative effects led to the establishment of more structured regulations and control over drug uses.

Nevertheless, Thalidomide is considered to be an effective therapy for the managements of Major RAS but because of its teratogenicity, it should be 53 restricted to male patients or females who have had hysterectomies or bilateral tubal ligations. In addition, Thalidomide could cause peripheral neuropathy which requires further investigation of Nerve Conduction Studies (NCS) and Electromyography (EMG) every 6 months to rule out this serious adverse event.

In a randomised, cross-over, multi-centre, placebo-controlled clinical trial, 100mg/day of Thalidomide was administered to 73 patients with severe RAS for 2 months. The patients were reviewed for more than 6 months. Although complete remission was obtained in 32 patients on Thalidomide and in 6 subjects on a placebo, recurrences were reported after discontinuation of the drug (Revus et al 1990). In a more recent multi-centre, retrospective study from 2003 to 2008 which was done by Hello et al in 2010 reported that Thalidomide achieved complete remission in 85% of 92 patients with severe RAS, within a median of 14 days. Despite its retrospective design, the data was collected from patients' medical notes and it was supplemented with patients' responses by phone interview, which made the accuracy of these interviews debatable.

In agreement with the previous studies, Mimura et al in 2009 demonstrated that when Thalidomide 100mg a day was administered to a total of 8 RAS patients, for a range of periods from 3-6 months (mean 4.8 months), 7 subjects (87.5%) had complete remission and 1 patient showed no response. One patient (12.5%) presented with drowsiness as side effect of Thalidomide. However, the sample size was very small and a further clinical trial with a larger sample size is needed to confirm their findings (Mimura et al 2009).

Furthermore, Letsinger et al in 2005 conducted an open retrospective study of 4 years follow-up which included 17 patients with complex aphthosis treated with Thalidomide, ranging from 50 to 150mg a day. The study found that 59% of patients achieved more than 50% improvement. Peripheral neuropathy as side effect of Thalidomide was noticed in 2 patients, which was confirmed with an Electromyography test (EMG). However, only 10 patients were available for follow-ups (Letsinger et al 2005). In a more recent study by Gil et al in 54

2010, a non-randomised, retrospective, open study revealed that remission was obtained in 47 patients with complex aphthosis who were treated with a low dose of Thalidomide (25mg/day); no adverse events were observed (Gil et al 2010).

In conclusion, Thalidomide demonstrated conclusive benefits in the treatment of complex aphthosis but caution must be considered due to its frequent adverse events. In addition, a major concern for the patients who were diagnosed with complex aphthosis is the possibility of progression to Behcet’s disease by 6-8 years (Verpilleux et al 1999). Therefore, close clinical follow- up and thorough investigations are necessary to monitor new signs and symptoms of Behcet’s disease.

1.6.8 Colchicine:

Colchicine is a toxic natural product of a plant called Colchicum autumnale which was first isolated in 1820 by French chemists PS Pelletier and J Caventon. It forms pale to green yellow crystals or powder when exposed to ultra violet radiation. It oxidizes into a dark colour, and must therefore be shielded from exposure to sunlight. Colchicine is rapidly absorbed when taken orally and reaches its peak plasma level in 30-120 minutes after ingestion; it is metabolized in the liver and distributed in the kidney and spleen. In addition, 50% of the drug circulates and links to plasma proteins and 20% of the dose eliminated in the urine (reviewed by Konda et al 2010).

Colchicine is an anti-inflammatory agent that limits leukocyte activity by binding to tubulin, a cellular microtubular protein, and therefore inhibiting protein polymerization (Sullivan et al 1998). It inhibits lysosomal degranulation and increases the level of cyclic AMP which decreases both the chemotactic and the phagocytic activity of neutrophils. Colchicine's anti-inflammatory action causes reductions in adhesiveness, mobility and chemotaxis of polymorphonuclear cells. It inhibits the T lymphocytes’ adhesions to the endothelial cells through interference with intercellular adhesion molecules

Selectins (Cronstein et al 1995). It reduces cellular secretion of procollagen and enhances collagenase production. Colchicine might have an immunosuppressive action as it inhibits cell-mediated immune responses by inhibiting IL-1 production and histamine release (Mekori et al 1989). Other pharmacological effects of this drug are the reduction of the corporal temperature, depression of respiratory centre, contraction of blood vessels, hypertension by central vasomotor stimulation and alteration of the neuromuscular function (review by Konda et al 2010).

Colchicine is used clinically for the treatment of gout, as a prophylaxis of recurrent gout and to prevent acute attacks by inhibiting leukocyte migration. Colchicine interferes with urate deposition by decreasing lactic acid production via leukocytes, inhibits kinin formation, and diminishes phagocytosis and the subsequent anti-inflammatory response. Colchicine has been used for the treatment of Familial Mediterranean Fever (FMF), Primary Biliary Cirrhosis (PBC), Behcet’s Disease, Epidermolysis Bullosa Acquisita, Dermatitis Herpetiformis, Chronic Bullous Dermatosis in children, Leukocytoclastic vasculitis and Urticarial vasculitis.

Adverse effects of Colchicine have been reported in the British National Formulary (BNF). Although Colchicine is generally well tolerated, the most frequent gastrointestinal adverse events include nausea, diarrhoea, vomiting, abdominal pain, gastritis and gastric bleeding. The gastrointestinal side effects are due to the increase in gut mobility by the inhibition of mitosis in the rapid turnover mucosa. However, these symptoms could be controlled by reducing the dose or using Intravenous administration (IV), which could minimise these side effects. Thus, Colchicine could be given as IV in 0.9% saline in a dose of 0.5mg and maximum of 2-3mg. However, Colchicine should not be given as Intramuscular (IM) or Subcutaneous (SC) as it might cause severe local irritation. In addition, long-term administration of oral Colchicine might cause bone marrow suppression, hepatic failure, renal impairment, alopecia, myopathy, peripheral neuropathy and blood disorders as it prevents the absorption of B12.

The interaction between Colchicine and Ciclosporin, Erythromycin, Statins, Calcium channel blockers, Itraconazole, Ketoconazole, Atazanavir, Digoxin and Grapefruit juice has been reported in the British National Formulary (BNF 2011). This interaction is which due to the fact that firstly; Colchicine is a P- glycoprotein substrate and all the others are P-glycoprotein inhibitors, which might cause an increase in the plasma level of Colchicine and raise the risk of Colchicine toxicity (Hansten et al 2011). Secondly; both Colchicine and the others are metabolised by the cytochrome P450 enzyme CYP3A4. Therefore concentrations of both could be increased by competition for metabolism and cause Colchicine toxicity, which may lead to myopathy and peripheral neuropathy especially in patients with renal impairment (Choi et al 1999).

Colchicine has been considered as a potent teratogen that should be avoided during pregnancy as it might pass into human milk; it should therefore be used with caution in nursing women. Despite that, Colchicine has been shown to be safe with a low dose of 0.5-1mg a day in children aged 3.5-11 years old with Periodic fever, Aphthous stomatitis, Pharyngitis and Adenitis disease (PFAPA), in a study with a 2 year follow up period where Colchicine managed to increase significantly the interval between episodes from 1.7 weeks to 8.4 weeks (p<0.006) (Tasher et al 2008).

1.6.8.a Use of Colchicine in the treatment of RAS:

Colchicine was first used for the treatment of RAS by Ruah et al in 1988. The study prescribed Colchicine for 3 patients with long standing active RAS and showed that all 3 patients experienced a significant reduction in symptoms, but recurrences occurred after 3 days discontinuation of the therapy. Diarrhoea as side effect of Colchicine was reported in one patient. Although the study recommended Colchicine treatment for Major RAS, the group sample size was very small (Ruah et al 1988). In a more recent study by Fontes et al in 2002, which included 54 patients with severe forms of RAS, subjects were prescribed 1.5mg of Colchicine a day for 3 months and followed for 4.7 years. The authors noticed a significant improvement of 57 ulcers in 63% of patients, where 12 subjects (22%) had complete remission during the treatment period and 22 patients (41%) improved significantly with a reduction of 50% in the frequency and the duration of the ulcers. Adverse events were reported in 10 patients (18.5%). Although this study followed the patients for a long period (4.7 years), all subjects were assessed over the telephone only.

In a study by Viguier et al in 2000 which included 5 cases of Herptiform RAS, it was shown that systemic steroids provided an effective cure in 2 patients while the preventive action of Colchicine was observed in 3 participants (Viguier et al 2000). The study results were promising but their sample size was very small and all the participants were men. In addition, an open, cross- over study which included 20 RAS patients with a high dose of Colchicine 1.5mg a day for 2 months was conducted by Katz et al in 1994. The study reported a significant reduction of 71% in the number of ulcers, alongside a reduction of 77% in the pain score, thus suggesting that Colchicine therapy has a major role in the prevention of RAS.

In a more recent study by Tappuni et al in 2008 which included 45 RAS patients who received Colchicine 500mcg a day combined with topical steroids (Betamethasone mouthwash) for 3 months, compared with 45 patients on Betamethasone mouthwash only. The study showed that Colchicine plus Betnesol mouthwash therapy significantly reduced number of ulcers, size, duration, site and pain, as well as improving the length of the ulcer free periods of RAS. The Betnesol mouthwash group as monotherapy noted a significant reduction in number of ulcers, size, duration, site and pain, but ulcer free periods had not improved (Tappuni et al 2008). These findings suggested that in the short-term topical steroids improved ulcer symptoms, while Colchicine seemed to prevent recurrences. This study has used a scoring system known as Ulcer Severity Score (USS) which converts RAS symptoms to numerical values to give each ulcer characteristic a score. These findings have been confirmed by the same authors in a more recent study where 79 RAS participants were included (41 Minor and 38 Major RAS) (Tappuni et al 2013). 58

In a randomised, double-blind clinical trial by Pakfetrat et al (2010), 34 RAS subjects were divided into two treatments groups with either a daily dose of Prednisolone 5m/day or Colchicine 0.5mg/day for 3 months. No significant differences were demonstrated between the two modalities of treatments in the number of ulcers, recurrences or pain. Due to the higher incidence of adverse events reported in Colchicine group (52.9%) compared to 11.8% in the Prednisolone group, the authors preferred Prednisolone for the management of RAS. However, the study did not consider the effect of long- term systemic steroids (Pakfetrat et al 2010).

A further study with a high dose of Colchicine (1.2mg/day for 2 months) was done by Hassan et al in 2010 which included 30 RAS patients. The study showed a significant reduction of 87% in aphthous counts and 82% in pain. Adverse events were reported during treatment periods, including (26.7%) abdominal pain, (13.3%) diarrhoea and (6.7%) vomiting. The study was carried over a short period of time with a high incidence of side effects (46.7%) (14 out of 30) which could be due to the high dose of Colchicine (Hassan et al 2010). A study which included 10 RAS patients taking a higher dose of Colchicine 1.5mg/day for 2-6 months was done by Mimura et al in 2009. The study revealed that 40% of patients showed no relapses during the treatment period (which was considered as complete remission), 50% reported reductions in the number of ulcers, duration and pain (which was considered as a moderate to mild response), while only 10% showed no response to Colchicine. Despite Colchicine demonstrating good results and being well tolerated by the participants, the sample size of 10 patients was very small which might affect their conclusions. In addition, the study was an open trial with the absence of blinding and randomisation, therefore the risk of bias cannot be ignored (Mimura et al 2009).

In contrast to the previous findings, Letsinger et al in 2005 demonstrated a 50% improvement in 11 out of 40 (27.5%) patients with complex aphthosis who received Colchicine 1.8mg/day as monotherapy, while the 12 out of 21 (57%) patients who received Colchicine and Dapsone achieved an 59 improvement of more than 50% in ulcers' symptoms (Letsinger et al 2005). Additionally, Lynde et al in 2009 conducted a retrospective study from 1998- 2007 which included 55 subjects with complex aphthosis. Out of the 50 patients who received Colchicine 1.8mg/day, 30 subjects (60%) achieved therapeutic success as monotherapy, where 29 had a 75% improvement and one patient had a complete remission during the treatment period. However, 13 (26%) patients had no response to the medication and 7 experienced side effects of diarrhoea (Lynde et al 2009). Despite the retrospective design of the previous study, a major concern should be applied in the treatment and management of complex aphthosis due to the fact that these patients could progress to Behcet’s disease by 6-8 years (Verpilleux et al 1999).

In conclusion, there was strong evidence in the majority of the studies that supported the efficacy and safety of Colchicine in the management of RAS. Colchicine demonstrated prevention of ulcer recurrences during treatment periods without the incidence of severe adverse events. However, the majority of the studies were for a short period and a large prospective, randomised clinical trial with a good timeframe would evaluate the effectiveness and safety of Colchicine in the management of RAS.

In summary, the methodological limitations in the previous studies were extensive. This impacted on the quality of their evidence along with the definition of RAS not always being specific, as recurrent oral ulcerations could be manifested as part of a broad spectrum of clinical disease ranging from Minor RAS to Behcet’s disease. In addition, utilising a standardised method for assessment of RAS that involves most of the ulcer characteristics such as the Ulcer Severity Score (USS) with structured management and specific follow-up periods may improve treatment outcome of RAS.

1.6.8.b Colchicine and its mode of action:

The anti-inflammatory effect of Colchicine is most likely mediated not only through direct interaction with microtubules but also through changes at the 60 transcriptional level in the acute attacks of Familial Mediterranean Fever (FMF). The latter effect apparently requires a higher concentration and takes a longer time to occur. This finding has been proven in a study that analysed the effect of Colchicine on global gene expression of Human Umbilical Vein Endothelial Cells (HUVEC). Thus the authors suggested that Colchicine does not have an immediate effect when prescribed (Ben-Chetrit et al 2006). A study by Kiraz et al in 1998, which included 11 patients with Familial Mediterranean Fever disease (FMF) compared with 10 healthy controls, noticed a significant decrease in serum concentrations of IL-6, IL-8 and TNF-a in patients who were tested before and after treatment with Colchicine for 2 months (Kiraz et al 1998).

In addition, Miller et al (1992) investigated Colchicine therapy and its effects on cytokines in the treatment of Primary Biliary Cirrhosis disease (PBC), in a study that included 28 patients with PBC who were tested before and during treatment with Colchicine compared with a placebo. The study demonstrated a significant reduction in serum levels of IL-2 and TNF-a during Colchicine treatment, while no changes were noticed in either IL-2 or TNF-a values in the placebo group (Miller et al 1992). Recently, Woo et al in 2012 showed that Colchicine significantly modulated the concentrations of TNF-a and IL-6 in peripheral blood mononuclear cells (PBMC) in patients with Behcet’s disease on Colchicine therapy. However, the differences in these two pro- inflammatory cytokines were seen in a responder group and healthy controls but not in a non-responder group (Woo et al 2012). In contrast, Manie et al in 1993 postulated that Colchicine failed to affect the expression of TNF-a and IL-6 in human monocytes (Manie et al 1993).

Sun et al in 2009 measured serum levels of IL-6, IL-8 and TNF-a in 43 subjects with the muco-cutanous type of Behcet’s disease. The study suggested that treatment with Levamisole and Colchicine for a period of 0.5- 11.5 months could significantly reduce serum concentrations of IL-6, IL-8 and TNF-a (p<0.001) (Sun et al 2009). In a more recent study with a large sample size of 70 children with FMF and anaemia were divided into three groups (group one; 17 patients newly diagnosed and untreated, group two; 36 61 patients on Colchicine therapy and group three; 17 healthy controls). The study showed that there were no significant changes in the levels of IL-6 between the first two groups (Celkan et al 2005). Although the authors suggested that Colchicine had a positive effect on disease activity and anaemia, the long-term prospects of Colchicine therapy in children (with its side effect of preventing the absorption of vitamin B12) are debatable.

In conclusion, previous studies have had contradictory findings regarding the effect of Colchicine on cytokines in different inflammatory disorders. Studies suggested that one of the factors involved in the pathogenesis of RAS was a cell-mediated immune response in which several cytokines seem to play a major role. Nevertheless, Colchicine was anticipated as having an anti- inflammatory mode of action by decreasing the mobility, adhesiveness and chemotaxis of polymorphonuclear cells as well as inhibiting T lymphocyte adhesions to the endothelial cells by interference with intercellular adhesion molecules Selectins (Cronstein et al 1995).

To date, there are no studies that have assessed the effect of Colchicine on pro-inflammatory and anti-inflammatory cytokines related to RAS in serum and saliva to confirm these theoretical findings.

1.6.9 Other treatments of RAS:

The use of intra-lesional injections in severe cases of RAS affecting quality of life such as eating, talking and swallowing, could be considered. Intra-lesional steroid injections which contain Betamethasone dipropionate and Betamethasone disodium phosphate have been used in a case report of a 30 year old female presenting with a history of 15 years with multiple Major RAS that affected her quality of life. Although there was a significant regression of the lesion after the first application, they needed to add systemic Levamisole to prevent recurrences (Picciani et al 2010). In addition, a single-blinded, placebo controlled trial using Botulinum toxin type A injection was performed on 70 RAS patients who randomly received either Botulinum or placebo 62 injections. Recurrences and adverse events were recorded for 6 days and Visual Analog Scale (VAS) was used to record pain (Yang et al 2009). Although a significant relief of pain after 3 days of the Botulinum injection was reported, the practical uses and the costs of this type of management are questionable.

Immediate relief of pain with a single session of Carbon dioxide (CO2) laser treatment was performed by Prasad et al in 2013 on 25 patients with Minor RAS. Pain levels were assessed immediately and after 24 hours using a numerical rating scale. Healing time was assessed on days 3 and 4 and once every 2 days thereafter for 2 weeks. The study showed an immediate relief of pain after treatment compared with pre-treatment (p<0.001) and a significant improvement in healing time (p<0.001) (Prasad et al 2013). This finding was supported by another study which included 20 Minor RAS patients (Zand et al 2012). Tezel et al in 2009 used Nd:YAG laser treatment on 20 RAS patients and demonstrated an immediate relief of pain and faster healing time (p<0.05).

In conclusion, the application, availability and practicality of these kinds of managements, especially in patients with other types of RAS like Herpetiform when there are 10-100 ulcers at a time, is debatable. (Table 7)

Table 7: Other types of RAS treatment

YEAR COUNTRY AUTHOR TREATMENT NO.pat RESULTS COMMENT

2009 Turkey Tezel Nd:YAG 23 Pain ↓ Randomise A duration ↓ controlled LASER

2010 Brazil Picciani Betamethasone 1 (F) No Case report BLS dipropionate recurrence 5mg & for 6 Betamethasone months disodium phosphate 2mg

(Injection)

2013 Oxford Prasad (CO 2) LASER 25 Pain ↓ (UK) RS Duration ↓ Placebo

2009 South Yang Botox injection 70 pain ↓ Single Korea TY no blinded Placebo recurrence controlled (injection) for 6 months 2012 Iran Zand (CO 2) LASER 10 duration ↓ Randomise N pain ↓ controlled placebo

In summary, current treatments for RAS are used with little or no clinical evidence for efficacy. There is preliminary evidence that Betnesol mouthwash and Colchicine has a role to play in the management of RAS. However, proper clinical trials have not been possible yet, partly due to the difficulty in measuring the outcome of such trials although recently a reproducible Ulcer Severity Score (USS) has been developed to measure the severity of the oral condition and therefore assist in monitoring the response to treatment. Future randomised clinical trials with a good timespan along with a precise definition of RAS and utilising a standardised method for assessment of RAS, which involves most of the ulcer characteristics such as the Ulcer Severity Score (USS), combined with structured management and specific follow-up periods, may improve treatment outcome of RAS.

Furthermore, relating the clinical changes to any change in specific immunological markers will provide new knowledge and will enhance the understanding of the mode of action of the medication used in RAS and their local and systemic therapeutic effect. To date, there were no studies which examine pro-inflammatory and anti-inflammatory cytokines related to RAS before and after treatment with Betamethasone mouthwash with and without systemic Colchicine in serum and saliva.

CHAPTER TWO MATERIALS AND METHODS

2.1 Clinical Trial Design:

This study is a randomised, prospective, parallel-group, single-centre clinical trial. The efficacy of three different modalities of treatment was assessed at 3 month intervals for 12 months on randomly selected Recurrent Aphthous Stomatitis (RAS) patients.

All subjects were recruited from referred patients to the Department of Oral Medicine at Guy’s Hospital for the diagnosis and management of RAS who had not received any form of topical or systemic therapy for their oral condition.

The trial was conducted in compliance with the principles of the Declaration of Helsinki, the principles of ICH-GCP and all of the applicable regulatory requirements. Research Ethics Committee approval No.06/Q0704/156 and regulatory approval from the MHRA were obtained. All subjects who met the inclusion criteria (see inclusion criteria section) were recruited after signing an informed consent form and given a patient information sheet. They were randomly assigned to one of the three treatments: Betnesol mouthwash, Colchicine tablet or Colchicine tablet and Betnesol mouthwash. The medications were prescribed using a prescription form specifically designed for the trial (Fig.2.1). The study was sponsored by Guy’s and St Thomas Foundation Trust, therefore the study medications were provided free of charge.

Fig.2.1

Pharmacy Department

Oral Medicine Study

Patient Name:……………… Patient ID No.:………

Date of Birth:…………… Hospital No.:…………

Please Tick (√) Dispensing Visit: 1 (month 0-3) ( ) Dispensing Visit: 2 (month 3-6) ( ) Dispensing Visit: 3 (month 6-9) ( ) Dispensing Visit: 4 (month 9-12) ( )

Please tick ( √) to supply medicines from one of the following groups:

Group 1: ( ) Betnesol® 500mcg Tablets: Dissolve ONE tablet in 10mls of water and use as a mouthwash for three minutes and then discard. Use QDS during ulcer attacks and BD in between Quantity to supply: 8 x 100 (6 months)

Group 2: ( ) Colchicine 500mcg Tablets: Take ONE tablet ONCE each day Quantity to supply: 2 x 100 (6 months)

Group 3: ( ) Colchicine 500mcg Tablets: Take ONE tablet ONCE each day Quantity to supply: 2 x 100 (6 months)

Betnesol® 500mcg Tablets: Dissolve ONE tablet in 10mls of water and use as a mouthwash for three minutes and then discard. Use QDS during ulcer attacks only Quantity to supply: 8 x 100 (6 months)

Prescribers Signature:……………… Date:……………………..

Print Name:…………………………. Contact:…………………….

For Pharmacy use Only

Dispensed By: Date:

Checked By: Date:

Prior to inclusion in the study, a full explanation of the trial and patient information sheet were given at least 24 hours before randomisation to enable the patient to consider participation, a full medical history was obtained and blood samples were taken for a Full Blood Count (FBC), Haematinics, Liver Function Test (LFT), Urea and Electrolytes (U&E).

On the screening visit, data was collected for each patient including age, sex and clinical characteristics of the ulcers. Patients included in the study were strictly Minor, Major or Herpetiform RAS as defined by Lehner 1968 (Fig.2.2). One of the criteria for exclusion were patients with recurrent oral ulcerations due to Haematological deficiencies, smoking cessation or presence of underlying systemic diseases including Behcet’s disease, Gastrointestinal diseases and Dermatological disorders (see exclusion criteria section).

Fig.2.2

Characteristics of the three types of Recurrent Aphthous Stomatitis

Major Minor Herpetiform

Number 1-10 2-5 10-100

Size (mm) >10 3-5 1-2 (but coalesce)

Peak age of onset 5-20 10-20 20-30 (years)

Duration 2-6 1-2 1-3 (weeks)

Scarring + - +/-

Mucosal site Keratinised& Non-keratinised Any site, especially non- keratinised floor of the mouth

Lehner (1968)

2.1.1 Inclusion criteria:

Patients were recruited into the study if they were: 1.Diagnosed with RAS as defined by Lehner 1968 (Fig.2.2). 2.Aged 18-65 years. 3.Subjects who were willing and able to give informed consent to participate in the study. 4.Patients who were not involved in other studies that would compromise their safety or undermine the scientific basis of the study.

2.1.2 Exclusion criteria:

Patients were excluded from the study if they were: 1.Under 18 years of age or over 65 years of age (due to the possible side effects of Colchicine and steroids in children and elderly patients such as adrenal suppression and gastric bleeding).

2.Unable to understand verbal explanations or written information given in English, or who have special communication needs and are unable to give informed consent (because difficulty in understanding the information and instructions required for the study would affect the study outcome).

3.Pregnant or breast feeding (because of the potential teratogenic effects of Colchicine and steroids such as adrenal suppression and congenital abnormalities).

4.Unwilling or unable to comply with the study protocol.

5.Patients with severe or relevant allergy history (because Colchicine may cause skin rashes as one of its side effects).

6.Patients with a history of hypersensitivity or side effects to any of the study medications.

7.On systemic steroids (to prevent interaction effects from other types of steroids which might affect the evaluation of the study outcome).

8.Diagnosed with systemic diseases such as Behcet’s disease, Crohn's disease, haematological deficiencies, gastrointestinal or dermatological disorders (to exclude oral ulcers with underlying systemic diseases).

9.On Colchicine or Betamethasone mouthwash in the previous three months.

10.Patients with blood dyscrasias or abnormal blood results.

11.Patients with a current or past history of cardiac, renal, hepatic or gastrointestinal disease (to prevent Colchicine's side effects of gastric bleeding, renal and hepatic impairment).

12.Patients who are debilitated in any way.

2.1.3 Randomisation:

A table of randomised numbers was used in this study in the sequence of “ABC” where (A) is Colchicine tablet and Betnesol mouthwash, (B) is Colchicine tablet and (C) is Betnesol mouthwash (Fig.2.3). In addition, a research nurse was in charge of the participants' randomisation, their assignments to the study groups and their review appointments. However, the sequence was concealed until the effect of all medications were analysed.

Fig.2.3 Randomisation table

1B B A C A C A B C B C A

2B A C C B A C B A C B A

3A C B C A B B C A C A B

4B B A A C C B C A B A C

5C C B B A A B A B C A C

6C C A B A B C B A B C A

7B C C A A B A C C B B A

8C C A B A B C B C A B A

9B A A C C B B C A A B C

2.1.4 Trial Medications:

All subjects were assessed by the same clinicians on all visits. The clinician was informed of which medication the participant was taking to ensure compliance. However, the assessment of the efficacy of treatment at each visit was carried out independently of the findings at the previous visits, to minimise bias. Neither the clinician nor the participants could be blinded because of the differences in the mode of the applications as one is a tablet and the other is a mouthwash.

The study medications are modalities of treatment which are widely used in Oral Medicine clinics for the treatment of RAS and the dosages used were the same as those routinely used for the management of this condition. These were:

1-Betnesol mouthwash was formulated by dissolving one tablet of Betamethasone sodium phosphate 500mcg in 10 ml of water and used as a mouth rinse for 3 minutes then discarded. This preparation was used four times a day when ulcers were present and twice a day in between ulcer attacks.

2-Colchicine tablet 500mcg once a day after breakfast.

3-Colchicine tablet 500mcg once a day and Betamethasone sodium phosphate 500mcg mouthwash as described above four times a day when ulcers are present only.

Patients continued on these medications for the duration of the study of 12 months. Potential side effects of using Colchicine tablets and Betnesol mouthwash as listed in the British National Formulary (BNF) were fully explained in the Patient's Information Sheet, thus participants could understand any potential risks before consenting to take part in the study. In addition, a direct phone number was given for advice and to contact one of the clinicians to report any adverse effects.

2.1.5 Clinical study design:

Initially, two hundred and one (201) patients were screened for the entry in the clinical trial according to the inclusion and exclusion criteria. Ninety-five (95) patients were excluded from the study due to either having atypical ulcers, abnormal blood results or systemic diseases. One hundred and six (106) RAS patients, comprised of 46 Males (age range of 21-63 with an age mean of 39.2 years) and 60 Females (age range of 22-65 with an age mean 72 of 40 years), who met the study inclusion criteria were randomised into one of the three study groups. Eighty-six (86) patients completed the whole study for the full twelve months: 39 Males (age range 21-63 years) and 47 Females (age range 23-65 years).

The diagnosis of RAS was made by trained clinicians in the Oral Medicine department at Guy's Hospital, according to clinical examination, investigations and history. The severity of the ulcers was evaluated using the Ulcer Severity Score (USS) which was established in the department by Tappuni et al (2005, 2013) (Fig.2.4). Individual characteristics of the ulcers were assessed; size, numbers, duration, pain, site and ulcer free periods were recorded and scored at the initial clinical visit (baseline visit) to produce a baseline score. This was repeated at every subsequent visit to compare the scores and assess any change during and after treatment.

Fig.2.4

KCL DENTAL INSTITUTE DEPARTMENT OF ORAL MEDICINE ULCER SEVERITY SCORE (USS) Recurrent Aphthous Stomatitis

Name: First visit to the department: Yes No Patient on medication for RAS: Yes No Date: Name of RAS medication: Clinician: Duration on medication:

Size Ulcer Score Description of USS

Characteristic

Average Size of Score = average size of ulcers in mm ulcers (in mm) Maximum score = 20 2mm

Average Number of Score = average number of ulcers ulcers in a crop 3mm Maximum score = 20

Average Duration Score = number of ½ weeks of ulcers i.e. Half week (3 days) scores 1, one and 4mm a half week (10 days) scores 3. Maximum score = 10

Ulcer-free period Score = 10 minus the average ulcer-free (in weeks) period in weeks Maximum score = 10 (never free from ulcers). 6mm

Pain as perceived 1 for slight discomfort when ulcers are by the patient present, 10 for excruciating ulcers (On a scale of 0-10) interfering with eating and talking. 8mm Maximum score = 10

Mucosal Site Group 1 Score = total of sites affected Labial mucosa 1 for each site in group 1 (non- Buccal mucosa keratenised mucosa) Buccal Sulcus 2 for each site in group 2 (keratinised Soft palate Specialised, or oropharynx 10mm Maximum score = 10 Ventral of tongue Lateral of tongue Floor of mouth

Group 2 Hard palate Attached gingival 12mm Alveolar ridge Dorsum of tongue Tonsils Pillars of fauces Uvula

Evidence of scarring Yes No Total

Tappuni et al 2005, 2013 (Modified by Surab Alsahaf 2010)

2.1.6 Ulcer Severity Score (USS): (Fig.2.4)

The USS is a scoring system that converts RAS symptoms to numerical values and gives an overall score for the severity of the ulcers at that point in time. It provides a summated score of six parameters based on a range of clinical ulcer characteristics (size, number, duration, ulcer free periods, pain and site). In addition, USS aids in measuring the severity of the oral ulcers and assists in monitoring the response to treatment as the severity of the individual ulcer characteristics could be assessed separately along with the effect of the medication on these individual parameters. This scoring system was based on the grounds that more than 95% of ulcers episodes are less than 20mm in diameter, less than 20 in number, last for less than 5 weeks and recur in less than 10 weeks (Tappuni et al 2013). Additionally, RAS assessment in the current study was based on the patient’s history, clinical examination, Clinical Record Form (CRF) and patient's diary. (Fig.2.5)

The calculation of the scores was as follows: (quoted from Tappuni et al 2013)

1. Size of ulcers: the score should be equal to the average diameter of the ulcers in millimetres (if the patient has ulcers with average size of 2mm; the score should be 2). The maximum score for this parameter was 20 to accommodate the Major type of RAS where ulcer size could be more than 12mm in diameter. There were diagrams of different diameter circles in the USS where patients could indicate their average ulcers diameter per episode.

2. Number of ulcers : the score should be equivalent to the average number of ulcers per episode (if the patient has on average 2 ulcers; the score should be 2). The maximum score for this parameter was 20 to include ulcer attacks that have more than 10 ulcers at a time, as seen in the Herpetiform type of RAS where a patient could have 10-100 ulcers at a time.

3. Duration of ulcers: the score for this parameter corresponds to the average duration of the ulcers calculated as one point for half a week (if the 75 patient has ulcers lasting for 10 days; the score should be 3). The maximum score was 10 to accommodate Major RAS which could last for 6 weeks.

4. Ulcer free period: the score for this parameter was 10 minus the average ulcer free period in weeks (if the patient was ulcer free for 2 weeks; the score should be 8). The maximum score was 10 to include patients who were never free from ulcers.

5. Pain: the score for this parameter was based on the patient's estimation on a scale of 0-10 where slight discomfort caused by the ulcers should score 1 while severe pain affecting patient's quality of life such as eating, talking and swallowing should score 10. Maximum score is 10.

6. Site : the score for this parameter was a combination of keratinised and non-keratinised mucosa as 1 point was given to non-keratinised mucosa (labial mucosa, buccal mucosa, buccal sulcus, soft palate, ventral surface of the tongue, lateral border of the tongue and floor of the mouth) and 2 points were given to keratinised and specialised mucosa (hard palate, attached gingiva, alveolar ridge, dorsum of the tongue, tonsils, pillars of fauces, uvula and oropharynx). The maximum score was 10.

Fig.2.5

Patient Daily Diary During Ulcer Attacks

2 mm 3 mm 4 mm 6 mm 8 mm 10 mm 12 mm Size of each ulcer (in mm)

Number of each ulcer

Pain of each ulcer ( 0 – 10)

Site of each ulcer

Inside right cheek Inside left cheek Inside upper lip Inside lower lip Floor of mouth Side of tongue Under tongue Back of palate On top of tongue Tonsils Palate Gingiva (gums)

Designed fot the study by Surab Alsahaf 2007

2.1.7 Study Subjects:

The study participants were randomised into 3 groups:

1. Group one: 35 Subjects received Betamethasone sodium phosphate (500 microgram tablets dissolved in 10ml of water and used as a mouthwash for 3 minutes then discarded) (Challacombe and Shirlaw 1991) administered four times a day when ulcers are present and twice a day in between ulcer attacks.

2. Group two: 35 Subjects received Colchicine tablets 500 micrograms a day.

3. Group three: 36 Subjects received Colchicine tablets 500 microgram a day and Betamethasone sodium phosphate 500 microgram tablets dissolved in 10ml of water and used as a mouthwash for 3 minutes then discarded, four times a day during ulcer attacks only.

2.1.8 Withdrawal of Subjects:

Participants were withdrawn from the study if: 1. They withdrew consent. 2. They did not comply with the study requirements. 3. They developed an adverse reaction to the medication.

Patients’ medical needs override their participation in the clinical trial and they were withdrawn if they required medical managements with systemic steroids or other treatments that interact with the medications used in the trial but their data was included in the study up to the point of their withdrawal. However, their participation was entirely voluntary and the clinical managements were not affected by declining to take part in the study.

2.1.9 Assessment of Safety:

1. Blood samples for Full Blood Count (FBC), Haematinics, Liver Function Test (LFT), Urea and Electrolytes (U&E) were taken and analysed at each scheduled visit. 2. Adverse events were recorded at each visit and subjects were given a direct telephone number to call and report any adverse reaction.

Furthermore, the trial was conducted in compliance with the principles of Declarations of Helsinki (1997), the principles of Good Clinical Practice (GCP) and all the applicable regulatory requirements. The study protocol and other documentations like amendments were submitted to the NHS Research Ethics Committee (REC) at Guy's Hospital. The investigators complied with the principles of GCP and the Medicines for Human Use (Clinical Trial Regulations 2004) with regards to reporting of SUSARs, AEs and SAEs, as well as providing the MHRA and the REC with annual progress reports, a final study report and monitoring regulatory inspections as requested.

2.1.10 Assessment of Efficacy:

1. Primary efficacy parameters: The primary outcome measure was the improvement of the USS as recorded on the final visit compared with the first visit basic score (baseline visit).

2. Secondary efficacy parameters: The secondary outcome measure was any change in the level of specific cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g) before and after treatment in serum and saliva samples which were collected at each visit from the participants.

2.1.11 Method

The patient group consisted of 106 participants (46 Males, 60 Females) with an age range of 18-65 years and age mean of 39.8 years, diagnosed with Minor and Major RAS, presented to the Oral Medicine department at Guy's Hospital (Guy’s and St. Thomas’ Hospital NHS Trust). Eighty-six patients finished the 12 month study period (39 Males, age range 21-63 years and 47 Females, age range 23-65 years). Twenty participants withdrew, of which 14 were due to difficulties in attending reviews appointments and 6 patients developed adverse events which necessitated termination of their medications.

Thirty-five RAS patients (11 Males, 24 Females) (age range 22-65) were randomised in the Betnesol group but 29 participants finished the 12 months trial (9 Males and 20 Females, age range 22-65 years) (17 Minor and 12 Major RAS). Twenty-six participants out of 35 in the Colchicine group completed the 12 months clinical trial (14 Males and 12 Females, age range 24-60 years) (18 Minor and 8 Major RAS). Thirty-one patients out of 36 in the Colchicine plus Betnesol group (16 Males and 15 Females, age range 21-63 years) (11 Minor and 20 Major RAS) completed the 12 months clinical trial.

2.1.12 Patients Evaluation:

All patients were reviewed and assessed every 3 months (a total of 5 visits) and the following was performed at each visit: (see activity sheet Fig.2.6)

1. History taken.

2. Extra-Oral and Intra-Oral examination.

3. Assessment of the severity of the RAS by the Ulcer Severity Score (USS), which includes size, number, duration, pain, site and ulcer free periods. (Fig.2.4)

4. The effects of the medication were assessed by analysing the patient’s daily diary which recorded the ulcers’ size, number, pain and site. (Fig.2.5)

5. Patients were asked to record any side effects and if the medication had been taken properly according to the instructions.

6. Blood tests: Full Blood Count (FBC), Haematinics, Liver Function Test (LFT), Urea and Electrolytes (U&E) were performed every 3 months to ensure that any changes before and after treatment were recorded.

7. Blood and saliva were taken at every visit to be analysed for the trial; centrifuged at 3000 revolutions for 10 minutes and stored at –70° C in the Oral Pathology department at Guy's hospital.

Fig.2.6 Activity Sheet for each visit

Screening visit Baseline visit Visit 1 Visit 2 Visit 3 Visit 4 Activities (before Tx) (3 months Tx) (6 monbth Tx) (9 months Tx) (12 months Tx) Full explanation of the trail Patient information sheet given Written informed consent Gp/GDP letter sent Demographics Medical history Medication history Social history Oral examination USS score Give/check USS diary and explain use Blood test Mouth swab / smear Saliva sample Blood results checked Saliva test result Dispense study treatment Confirm treatment had been taken Check for RAS Tx adverse effects Give date for next visit Discharge to oral med clinic/GP/GDP Total estimated time needed 50 minutes 45 minutes 35 minutes 35 minutes 35 minutes 35 minutes Comments

2.1.13 Trial Statistics:

The statistical power of this study was based on a retrospective observational analysis of 79 patients’ notes of RAS subjects on topical steroids which showed a range of outcome measure change in the Ulcer Severity Score (USS) of 0-51 with a mean change of 27.4 and Standard Deviation (SD) of 11. This study was designed to have 80% power to detect differences of 5 (SD=10) among the three groups in terms of mean Ulcer Severity Score. At this level of power, it was calculated that up to 64 patients per group would be ideal, but in the current study 35 patients per group were obtained due to the fact that the majority of the patients were on Betnesol mouthwash before their referral which is one of the exclusion criteria.

Paired t-Test was performed to assess response to treatments within each group of the study. The null hypothesis was set as subjects taking treatment would report a significant improvement, which was set as p<0.05, in the Ulcer Severity Score (USS) (Fig.2.4) compared in each visit with the baseline and with the previous review where mean and percentage values were conducted in this clinical trial. In addition, unpaired t-Test was performed to assess cytokine concentrations in RAS subjects and healthy controls. All data were checked on regular basis (every 6 months) by the Medicines and Healthcare Products Regulatory Agency (MHRA) investigators at Guy's hospital .

2.2 Laboratory Study Design:

Relating the clinical changes to any change in specific immunological markers (cytokines) would enhance the understanding of the mode of action of the medications used in this trial and their local and systemic therapeutic effects. Therefore, the secondary outcome measure in this study was to observe any changes in the level of nine cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g), before and after treatment in serum and salivary samples of RAS patients compared with healthy controls using the Luminex Multi-

Analyte Profiling System (MAP). Serum and salivary samples were collected from 51 healthy controls aged-matched with a mean age of 40.7 years (27 Females, age range 21-65 years and 24 Males, age range 26-64 years) compared with 72 RAS patients (39 Females, 33 Males) classified as: 25 subjects in the Betnesol group (age range 22-65 years), 22 subjects in the Colchicine group (age range 24-60 years) and 25 in the Colchicine plus Betnesol group (age range 21-63 years). All samples were collected before treatment (baseline visit) and after treatment (final visit) and were analysed using the Luminex Multi-Analyte Profiling System (MAP). The advantage of Luminex technology lies in its significant reductions in time and cost compared with Enzyme-linked immunosorbent Assay (ELISA), especially when many immunological markers needed to be measured and sample volumes were limited (Arellano-Garcia et al 2008).

2.2.1 Serum samples:

After obtaining written consent and provision of a detailed Information Sheet at least 24 hours before the baseline visit to enable the patient to consider participation, venous blood samples were collected by an aseptic technique. The antecubital area was chosen for the blood collection and was prepared by applying an antiseptic solution. A tourniquet was applied around the arm and a butterfly kit was used to obtain blood (5ml) from the participants into a yellow-top tube then centrifuged at 3000 revolutions for 10 minutes and stored at –70° C in the Oral Pathology department at Guy's hospital.

2.2.2 Saliva samples:

Whole un-stimulated saliva (5ml) was collected at every visit in sterile universal containers by asking the participant to spit into the container for 10 minutes then centrifuged at 3000 revolutions for 10 minutes and stored at – 70° C in the Oral Pathology department at Guy's hospital.

2.2.3 Laboratory procedure:

Fluorokine Multi-Analyte Profiling systems (Fluorokine MAP) Human base kit A was used according to R&D systems. Saliva and serum samples were diluted four times with calibrator diluents for the determination of nine cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g). Initially, the 96-well filter bottom plate was pre-wet with a wash buffer. Fifty µl of diluted microparticle solution and 50 µl of each sample were added to each well in duplicate, then the plate was incubated for 3 hours and washed three times with a wash buffer. Afterwards, 50 µl of diluted biotin antibody was added to each well and incubated for 1 hour. The plate was then washed three times with a wash buffer and 50 µl of diluted Streptavidin-PE added to each well and incubated for 30 minutes. All incubations were performed at room temperature on a horizontal shaker. Samples were analysed in duplicate and the concentrations calculated according to the Standards. Finally, the plate was washed with 100 µl of wash buffer and read using the Luminex analyzer within 90 minutes. The assays were run on a 96-well plate format, followed by detection on a Luminex 100 instrument. As the beads run through the instrument, the internal dyes are excited by a laser, which results in the classification of each bead. Another laser excites the reporter dye which is directly proportional to the amount of analyte bound to each bead. The resulting fluorescence is recorded by the instrument which then provides the median fluorescence unit obtained from measuring 100 beads (see section 2.2.3).

2.2.4 Assay Procedure: (Quoted from the company booklet)

1. The filter bottomed micro-plate was pre-wet by filling each well with 100 µl of wash buffer and then the liquid was removed through the filter at the bottom of the plate using a vacuum.

2. Fifty µl of microparticle mixture was added to each well then 50 µl of the sample was added to each well and incubated for 3 hours on a horizontal 84 shaker at room temperature (it was covered with foil to protect from light all the time).

3. Plate was then washed by removing the liquid from each well using a vacuum then each well filled with 100 µl of wash buffer before vacuuming again. The process was repeated 3 times.

4. 50 µl of diluted biotin antibody cocktail was added to each well, covered with foil to protect from light and then incubated for 1 hour on a horizontal shaker at room temperature.

5. 100 µl of wash buffer was used to wash the plate 3 times (as described above), 50 µl of Streptavidin-PE was added to each well and was incubated for 30 minutes on the horizontal shaker at room temperature.

6. 100 µl of wash buffer was used to wash the plate in the same process as described above and then incubated for 2 minutes on the horizontal shaker at room temperature and covered with foil.

7. The plate was read within 90 minutes using the Luminex analyzer which normally requires more than one hour.

2.2.5 Luminex Multi-Analyte Profiling System (MAP): (Quoted from the company website)

The Luminex system is a combination of flow-cytometry and fluorescent microspheres into multi-analyte profiling technology. The core xMAP technologies that make up the Luminex System are the xMAP microspheres and the Luminex analyzer. The xMAP microspheres are a family of 100 fluorescently dyed 5.6 micron polystyrene microspheres, which serves both as a spectral identifier and a solid surface to build analyte-specific assays. (Fig.2.7)

Fig.2.7

The xMAP detection system can identify which set the microsphere belongs to based on the unique spectral signature achieved by internally filling each of the 100 bead sets with a known ratio of red and infrared fluorophores. These colour-coded microspheres are then conjugated to reagents (e.g., antibodies, receptors, peptides) allowing the capture and detection of up to 100 specific analytes from a heterogeneous sample.

The Luminex analyzer is a flow-cytometry based instrument that combines fluidics, 2 lasers, 4 detectors, and real-time digital signal processing to distinguish the 100 different sets of colour-coded microspheres, each bearing an analyte-specific assay. The fluidics system of the Luminex analyzer aligns the microspheres into a single file as they enter a stream of sheath fluid and then enter a flow cell. (Fig.2.8)

Fig.2.8

Once the beads are in single file within the flow cell, each bead is individually interrogated by the red 635 nm laser which is used to excite the dyes inside the beads to determine their "colour" or "region" and also to discriminate doublets by light scatter. The green 532 nm assay laser is used to excite phycoerythrin (PE), which is the reporter dye captured during the assay. (Fig.2.9)

Fig.2.9

The fluorescent signal generated from the lasers exciting both the dyes within the microspheres and the PE reporter dye molecules are distinguished by 4 detectors within the Luminex analyzer. One detector measures side scatter to size the microspheres and remove signals related to other materials passing through (typically air bubbles or microsphere aggregates). A second measures the internal red dye concentration while the third measures the internal inferred dye concentration. The integration of this data defines and identifies the bead. A photo multiplier tube (PMT) associated with the green laser then measures the reporter signal for the event based on size and dye content. Depending on the specific protein or nucleic acid-based bioassay, the reporter signal intensity can be used to qualitatively identify a biomarker and/or quantitatively determine protein levels in a clinical sample. (Fig.2.10)

Fig.2.10

2.2.6 Reproducibility:

Nine cytokines levels were measured in serum and salivary samples taken from 72 RAS patients and 34 healthy controls using the Luminex Multi- Analyte Profiling System (MAP) as described above. To assess the reproducibility of these measurements, two additional replicate assays were run on three different days for inter-reproducibility and on one day for intra- reproducibility with a Coefficient of Variation (CV) ranging from 0-37% in saliva and 8-30% in serum for the inter-reproducibility (Table.8 and Table.9). For the intra-reproducibility, CV ranged from 0-50% in saliva and 0-27% in serum (Table.10 and Table.11).

(Table.8) Inter-Reproducibility of nine cytokines (saliva) saliva IL-8 IFN-g IL-4 Il-6 IL-10 IL-17 TNFa IL-2 IL-5 2082 4 4 7.5 3 6.5 4 2.75 1 2031.75 3.5 5 6 3 7 4 2.5 0.5 1718 3 4 6.5 3 6.5 3 3 1

Mean 1943.9 3.5 4.3 6.7 3 6.7 3.7 2.8 0.8 SD 197.3 0.5 0.6 0.8 0 0.3 0.6 0.3 0.3 SE 113.9 0.3 0.3 0.5 0 0.2 0.3 0.2 0.2 CV 10% 14% 13% 11% 0% 4% 16% 10% 37%

(Table.9) Inter-Reproducibility of nine cytokines (serum) serum IL-8 IFN-g IL-4 Il-6 IL-10 IL-17 TNFa IL-2 IL-5 19.25 2.5 2.5 1.25 2.5 4.5 2.25 2 1 20.5 3.5 3.75 1.5 4 6.5 3.5 3 1.5 17.3 2.8 2.5 1 3 4.5 2 2.5 1

Mean 19 2.9 2.9 1.3 3.2 5.2 2.6 2.5 1.2 SD 1.6 0.5 0.7 0.3 0.8 1.2 0.8 0.5 0.3 SE 0.9 0.3 0.4 0.2 0.5 0.7 0.5 0.3 0.2 CV 8% 17% 24% 23% 25% 23% 30% 20% 25%

(Table.10) Intra-Reproducibility of nine cytokines (saliva)

saliva IL-8 IFN-g IL-4 Il-6 IL-10 IL-17 TNFa IL-2 IL-5 2482 4 4 7.5 3 6.5 4 2.75 1 2031.75 3.5 5 6 3 7 4 2.5 0.5

Mean 2256.9 3.8 4.5 6.8 3 6.8 4 2.6 0.8 SD 318.4 0.4 0.7 1.1 0 0.4 0 0.2 0.4 SE 225.1 0.3 0.5 0.8 0 0.3 0 0.1 0.3 CV 14% 10% 15% 16% 0% 5% 0% 7% 50%

(Table.11) Intra-Reproducibility of nine cytokines (serum)

serum IL-8 IFN-g IL-4 Il-6 IL-10 IL-17 TNFa IL-2 IL-5 30.5 3.15 3.25 1.5 4.2 5.5 3.5 3 1.5 20.5 3.5 3.75 1.5 4 6.5 3.5 3 1.5

Mean 25.5 3.3 3.5 1.5 4.1 6 3.5 3 1.5 SD 7.07 0.24 0.35 0 0.14 0.7 0 0 0 SE 5 0.1 0.2 0 0.1 0.5 0 0 0 CV 27% 7% 10% 0% 3% 11% 0% 0% 0%

CHAPTER THREE RESULTS Comparison of three different modalities of treatments for the management of Recurrent Aphthous Stomatitis (RAS)

3.1 Complete series:

106 RAS patients who met the inclusion criteria were randomised into the three study groups (see Materials and Methods). The 86 subjects who completed the clinical trial were followed for 12 months at 3 monthly intervals (a total of 5 visits), 20 participants withdrew because of work commitments and 6 subjects had adverse events which necessitated termination of their treatment (Fig.3.1). Of the 86 RAS subjects who were diagnosed according to Lehner's criteria (Lehner 1968), 46 had Minor RAS with an Ulcer Severity Score (USS) range of 20-37 and 40 had Major RAS with an USS range of 33- 53.

Demographic distribution of trial patients

Side effect 6

Withdraw 20

Finished 86

Randomised 106

0 20 40 60 80 100 120

Fig.3.1 Distribution of the randomised 106 RAS patients. 86 completed the 12 month follow ups, 20 withdrew, 6 of which were due to side effects.

The mean Ulcer Severity Score (USS) for the 86 RAS patients before treatment (baseline visit) was 34.9 ± 7.2 (SD). This significantly decreased after 3 months (visit 1) to a mean of 24.5 ± 9.1, representing a significant improvement of 29.8% (p<0.001). After 6 months (visit 2) the USS mean was

21.1 ± 9.2, giving a mean improvement of 39.5% (p<0.001). After 9 months (visit 3) the USS mean was 18.8 ± 10.4, giving a significant improvement of 46.1% (p<0.01). Although there was a small further reduction in the mean USS to 17.5 ± 8.9, with a further improvement of 49.9% after 12 months, this was not statistically significant (p<0.1) compared with the 9 months visit but remained significantly less than baseline visit (p<0.0001). (Fig.3.2)

86 RAS patients followed for 12 months 40 29.8% 39.5% 35 46.1% 49.9% 30 * 25 * * USS 20 15 10 5 0 baseline 3 months 6 months 9 months 12 months USS 34.9 24.5 21.1 18.8 17.5

Fig.3.2 RAS patients (n=86) followed for 12 months at 3 month intervals (a total of 5 visits). Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test *p<0.05. ±SEM

Of the 86 RAS patients, there were 46 Minor RAS subjects in this study, with a range of USS before treatment of 20-37 and a mean of 29.5 ± 3.9 (SD). After 3 months therapy the mean USS decreased to 21.4 ± 7.4, which gave an average improvement of 27.4% (p<0.001). After 6 months there was a significant reduction to 17.9 ± 8.6, with a mean improvement of 39.3% (p<0.001). A further significant reduction was seen at the 9 months follow up with a mean of 15.8 ± 9.6, and a mean improvement of 46.4% (p<0.04). Although there was a small further reduction of the mean USS to 14.5 ± 8.5, with a further improvement of 50.8% after 12 months therapy, this was not statistically significant compared with the previous follow up of 9 months (visit 3) (p<0.07) but remained significantly less compared with the baseline visit (p<0.001) . (Fig.3.3)

46 Minor RAS patients followed for 12 months 35 27.4% 39.3% 46.4% 50.8% 30 * 25 * * 20

15 USS 10

0 baseline 3 months 6 months 9 months 12months USS 29.5 21.4 17.9 15.8 14.5

Fig.3.3 Minor RAS (n=46) patients reviewed at 3 monthly intervals for 12 months. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.05. ±SEM

There were 40 Major RAS subjects in this study, with a USS range before treatment of 33-53 and a mean of 40.8 ± 5.2 (SD). After 3 months therapy the mean USS decreased to 27.5 ± 9.9, giving a significant mean improvement of 32.6% (p<0.001). After 6 months a further reduction of the mean USS to 24.3 ± 8.6 was seen, with a significant mean improvement of 40.4% (p<0.006). A further significant reduction in the mean USS to 22.1 ± 10.4 was seen at the 9 months follow up, giving a mean improvement of 45.8% (p<0.02). Although there was a small further reduction of the mean USS to 21 ± 8.1 with a further improvement of 48.5% after 12 months therapy, this was not statistically significant compared with the 9 months reviews (visit 3) (p<0.1) but remained significantly less compared to the baseline visit (p<0.001). (Fig.3.4)

40 Major RAS patients followed for 12 months 45 40 32.6% 40.4% 45.8% 48.5% 35 * 30 * * 25 20 USS 15 10 5 0 baseline 3 months 6 months 9 months 12months USS 40.8 27.5 24.3 22.1 21

Fig.3.4 Major RAS (n=40) patients reviewed at 3 monthly intervals for 12 months. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.05. ±SEM

Six ulcer characteristics (size, number, duration, pain and site of ulcers) and the length of ulcer free periods make up the USS. Each of these characteristics was examined before and after treatment to determine whether treatments might have an effect on each of these ulcer parameters. At the baseline visit the mean size score for the 86 RAS subjects was 5.4 ± 2.4 (SD) (range 2-14), the mean score for the number of the ulcers present was 3.4 ± 1.2 (range 1-6), the duration of the ulcers mean score was 4 ± 2 (range 2-10), the mean score of the ulcer free periods was 7.6 ± 1.9 (range 2- 10) and an average sites affected score of 7.4 ± 2.3 (range 3-10). Patients gave an average pain score of 6.9 ± 1.7 with a range of 3-10. (Fig.3.5)

After 6 months of treatment significant reductions were found in all six characteristics of size, number, duration of ulcers, length of ulcer free periods, pain and sites affected. The greatest improvements were seen in number, size and duration of ulcers. The mean improvements (reductions) were as follows: 45.9% in size, 51.9% in number, 49.4% in duration of the ulcers, 38.6% in pain, 41.3% in site as well as an increase in the length of ulcer-free periods by 25.4% (p<0.001). After 12 months further significant reductions were noticed in the size of the ulcers by 53.5%, duration by 57.1%, site by

47.6%, pain by 48.2% and there was an increase in ulcer-free periods by 42.6% (p<0.05). Improvement in the number of ulcers by 56% was not statistically significant compared with scores at 6 months (p<0.1). (Fig.3.5)

86 patients followed for 12 months

9 45.9% 51.9% 49.4% 25.4% 38.6% 41.3% 8 7 53.5% 56% 57.1% 42.6% 48.2% 47.6% 6 5 4 baseline 3 2 6 months 1 12 months 0 size no duration ulcer free pain site baseline 5.4 3.41 4.03 7.63 6.91 7.36 6 months 2.92 1.64 2.04 5.69 4.24 4.32 12 months 2.51 1.5 1.73 4.38 3.58 3.86

Fig.3.5 Comparison of mean scores of individual ulcer characteristics (n=86 RAS patients) at baseline visit, 6 and 12 months reviews. Mean % improvement compared with the baseline visit using paired t-test. ±SEM

3.2 USS related to treatment:

3.2.1 (Group1) Betnesol mouthwash: Betamethasone phosphate (500mcg) dissolved in 10 ml of water and used as a mouthwash four times a day (QDS) during ulcer attacks and twice a day (BD) in between.

A total of 35 RAS patients who met the inclusion criteria were enrolled in this group in which they received Betnesol mouthwash QDS during ulcer attacks and BD in between as monotherapy. Six patients (17.1%) (4 Females, 2 Males) withdrew due to work and family commitments and the data for these patients was not used. Therefore, this study was effectively conducted for 29 patients (age range 22-65 years, mean age 41.2 years). There were 20 Females (age range 23-65, mean age 41.5) and 9 Males (age range 22-63 years, mean age 40.7 years). Seventeen subjects were presented with Minor RAS while 12 were presented with Major RAS. Of the 29 patients who were

95 reviewed at 3 monthly intervals for 12 months, 4 patients (13.8%) showed no recurrences during the treatment period, 22 patients (75.9%) reported greater than 20% improvement in their USS, while 3 subjects (10.3%) showed no change or a slightly higher USS. None of the patients reported side effects during the trial period. (Fig.3.6)

Demographic distribution of RAS patients in Betnesol group

side effect 0

withdraw 6

Finish 29

Randomise 35

0 10 20 30 40

Randomise Finish withdraw side effect RAS 35 29 6 0

Fig.3.6 Distribution of RAS patients in the Betnesol group, 35 were randomised, 29 completed the 12 month follow ups, 6 withdrew and no side effects were reported.

3.2.1.a Ulcer Severity Scores (USS) of 29 RAS patients before and after treatment with Betnesol mouthwash:

The range of the USS for the 29 RAS subjects before treatment (baseline visit) with Betnesol mouthwash was 22-53 with a mean of 34.6 ± 7.9 (SD). This was significantly decreased after 3 months therapy (visit 1) to a mean of 22.5 ± 8.1, (range 0-35), giving a mean improvement of 35% (p<0.0001). After 6 months (visit 2) there was a further reduction of the mean USS to 18.3 ± 8.3, giving a 47.1% improvement (p<0.0005). After 9 months (visit 3) a further reduction of the mean USS to 15.3 ± 9.6 was seen, giving a significant improvement of 55.8% (p<0.004). However, no further improvements were

96 noticed after 12 months treatment (visit 4) (range 0-30, mean 15.7 ± 9.3) (SD) with a mean USS improvement of 54.6% (p<0.4). (Fig.3.7)

29 patients on Betnesol mouthwash QDS/BD followed for 12 months 40 35 35% 47.1% 55.8% 54.6% 30 25 * * USS 20 * 15 10 5 0 baseline vit. 1 vit.2 vit. 3 vit. 4 USS 34.6 22.5 18.3 15.3 15.7

Fig.3.7 RAS patients (n=29) reviewed for 12 months on Betnesol mouthwash QDS/BD. Figures above columns show the USS mean % improvement compared with the baseline and significance compared with the previous visit using paired t-test * p<0.001. ±SEM

3.2.1.b Comparison of Ulcer Severity Scores (USS) in Minor and Major RAS after treatment with Betnesol mouthwashes:

There were 17 Minor RAS subjects in this group, with a range of USS before treatment of 22-36 with a mean of 29.2 ± 3.9 (SD). After 3 months therapy the USS ranged between 0-29 with a mean of 20.7 ± 7.3, giving an average improvement of 29.2% (p<0.0001). After 6 months the mean USS decreased to 17.4 ± 8, giving a mean improvement of 40.4% improvement (p<0.02). After 9 months there was a further significant reduction of the mean USS to 14.2 ± 9.7, giving a mean improvement of 51.5% (p<0.01). Although there was a small further reduction of the USS mean to 13.2 ± 9.4 (SD) after 12 months therapy with a further improvement of 54.7%, this was not statistically significant compared with the previous follow up of 9 months (visit 3) (p<0.1)

97 but remained significantly less when compared to the baseline visit . (Fig.3.8, Table 12)

17 Patients on Betnesol mouthwash QDS/BD followed for 12 months (Minor RAS) 35 29.2% 40.4% 51.5% 54.7% 30 25 * * 20 * USS 15 10 5 0 baseline 3 months 6 months 9 months 12 months USS 29.23 20.7 17.41 14.18 13.23

Fig.3.8 Minor RAS (n=17) patients reviewed at 3 monthly intervals for 12 months on Betnesol mouthwash QDS/BD. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.05. ±SEM

Analysis of the responses of individual patients showed considerable variation (Fig.3.8b). All of the patients with Minor RAS had a USS of greater than 20 at the beginning of the treatment period and 14 of the 17 patients had a USS of less than 20 after 9 months. Some patients showed a steady decline in ulcer severity over this time, while others showed a rapid improvement but then relapses. Little further improvement was found after 9 months, suggesting that this might be an optimal time to assess the efficacy of treatments. (Fig.3.8b)

pat 1 17 Patients on Betnesol mouthwash followed for 12 months ( Minor RAS) pat 2 pat 3 pat 4 40 pat 5 35 pat 6 pat 7 30 pat 8 25 pat 9 20 pat 10 pat 11 15 pat 12 10 pat 13 5 pat 14 0 pat 15 pat 16 1 2 3 4 5 pat 17 Fig.3.8b The effect of Betnesol QDS/BD on individual Minor RAS patients (n=17) reviewed for 12 months, visit 1 (baseline), visit 2 (3 months), visit 3 (6 months), visit 4 (9 months) and visit 5 (12 months).

There were 12 Major RAS subjects in this group with a USS range before treatment of 35-53 with a mean of 42.1 ± 5.5 (SD). After 3 months therapy a significant reduction of the USS was seen with a range of 0-35 and a mean of 25.1 ± 8.9, giving a mean improvement of 40.4% (p<0.0002). After 6 months a further reduction of the mean USS to 19.7 ± 8.8, was seen with a significant mean improvement of 53.1% (p<0.001). Although there was a further improvement of the USS mean to 16.8 ± 9.7, giving a mean improvement of 60.2% after 9 months, this was not significantly different from that at 6 months (p<0.1). After 12 months therapy there was a slight increase in the USS compared with 9 months (though very significantly reduced compared with the baseline) with a mean of 19.1 ± 8.4 (SD), giving a mean improvement of 54.7% (p<0.1). (Fig.3.9, Table 12)

12 Patients on Betnesol mouthwash QDS/ BD followed for 12 months (Major RAS)

50 40.4% 45 53.1% 60.2% 54.7% 40 35 30 * USS 25 * 20 15 10 5 0 baseline 3 months 6 months 9 months 12 months USS 42.08 25.08 19.72 16.75 19.08

Fig.3.9 Major RAS (n=12) patients reviewed at 3 monthly intervals for 12 months on Betnesol mouthwash QDS/BD. Figures above columns show the USS mean % improvement compared with the baseline and significance compared with the previous visit using paired t-test * p<0.001. ±SEM

Analysis of the responses of individual patients showed a considerable variation (Fig.3.9b) but perhaps less than for Minor RAS. All of the patients with Major RAS had an USS of greater than 30 at the beginning of the treatment period and by 6 months, all but one had an USS of less than 30, and 6 of the 12 patients had an USS of less than 20. Some patients showed a steady decline in the ulcer severity over this time, while others showed a rapid improvement but then relapses. One patient appeared to be recalcitrant to therapy, and in one subject ulcers virtually disappeared. Little further improvement in the group was found after 6 months, suggesting that this might be a minimal period to assess the efficacy of treatments (Fig.3.9b).

100

12 Patients on Betnesol mouthwash followed for 12 months Major RAS

pat 1 60 pat 2 50 pat 3 pat 4 40 pat 5 30 pat 6 pat 7 20 pat 8 10 pat 9 pat 10 0 pat 11 baseline vit. 1 vit.2 vit. 3 vit. 4 pat 12

Fig.3.9b The effect of Betnesol QDS/BD on individual Major RAS patients (n=12) reviewed for 12 months, visit 1 (baseline), visit 2 (3 months), visit 3 (6 months), visit 4 (9 months) and visit 5 (12 months).

Table.12 Summary of Ulcer Severity Scores before and after treatment with Betnesol mouthwash QDS/BD for 12 months in patients with Minor and Major RAS

Type of No. of USS USS USS USS mean% P value RAS patients range mean range mean improvement (compared before before after Rx after with Rx Rx Rx baseline)

Minor 17 22-36 29.2 0-28 13.2 54.7% P<0.0001 ± 3.9 ± 9.4 (SD) (SD)

Major 12 35-53 42.1 8-30 19.1 54.6% P<0.0001 ± 5.5 ±8.4 (SD) (SD)

101

3.2.1.c Effect of Betnesol mouthwash on six individual characteristics of ulcers:

The USS score is measured using: size, number, duration of ulcers, pain and sites affected as well as the length of ulcer free periods. Each of these characteristics was examined before and after treatment to determine whether treatment with a steroid mouthwash might have a differential effect. At the baseline visit, the mean size of ulcer score was 5.2 ± 2.5 (SD), the mean number of ulcers was 3.4 ± 1.3, the mean duration of ulcer score was 3.9 ± 1.7 and the mean ulcer free period was 7.8 ± 1.7 with an average sites affected score of 7.4 ± 2.4. Patients reported an average pain score of 6.8 ± 1.7 (SD). (Fig.3.10, Table 13)

The severity of each of the six ulcer characteristics scores were found to be significantly decreased over 6 then 12 months of treatment. There was a reduction in the size of the ulcer by 51.7%, number of ulcers by 57.9%, duration of ulcer by 59.3%, pain by 49.9%, site by 48.7% and an increase in ulcer free periods by 27.2% after 6 months therapy of Betnesol mouthwash (p<0.001). After 12 months treatment there was a significant decrease in the number of ulcers by 63.2% (p<0.03) and an increase in the ulcer free periods by 45.2% (p<0.01). However, there were no further significant reductions in size (55.3%) (p<0.3), duration of ulcer (64.5%) (p<0.2), pain (53.5%) (p<0.4) and site (56.5%) (p<0.1). (Fig.3.10, Table 13)

102

29 patients on Betnesol mouthwash followed for 12 months 9 27.2% 8 51.7% 57.9% 59.3% 49.9% 48.7% 7 45.2% 6 55.3 % 56.5% 5 64.5% 53.5 % 4 63.2% 3 2 baseline 1 6 months 0 ulcer 12 months size no duration pain site free baseline 5.24 3.37 3.86 7.79 6.82 7.44 6 months 2.53 1.42 1.57 5.67 3.42 3.82 12 months 2.34 1.24 1.37 4.27 3.17 3.24

Fig.3.10 Comparison of six ulcer characteristics scores (n=29 RAS patients) at baseline visit, 6 and 12 months reviews. Mean % improvement compared with the baseline visit using paired t-test. ±SEM

Table.13 Summary of ulcer characteristics scores before and after treatment with Betnesol mouthwash QDS/BD for 12 months

Ulcer Score (me an) Score (mean) % improvements P value characteristics before Rx± (SD) after Rx ± (SD) ( compared with baseline) Size 5.2 ± 2.5 2.3 ± 1.3 55.3 % P<0.001 No 3.4 ± 1.3 1.2 ± 0.8 63.2 % P<0.001 Duration 3.9 ± 1.7 1.4 ± 0.8 64.5 % P<0.001 Ulcer free 7.8 ± 1.7 4.3 ± 3.4 45.2 % P<0.001 Pain 6.8 ± 1.7 3.2 ± 2 53.5 % P<0.001 Site 7.4 ± 2.4 3.2 ± 2.9 56.5 % P<0.001

Overall clinical responses benefits : 13.8% of patients had no recurrences during treatment periods and 10.3% did not show any change. Thus 75.9% showed a reduction in their USS using Betnesol mouthwash QDS during ulcer attacks and BD in between, with no side effects reported for the duration of the follow up period of 12 months. In addition, patients noted the benefit within the first 3 months of therapy, with maximum statistically significant effects by

103

9 months (p<0.004). The effect of Betnesol on improving the USS was seen with Minor RAS for up to 9 months, while with Major RAS improvements were seen in the first 6 months (p<0.05). RAS patients on Betnesol mouthwash QDS and BD had a significant reduction in all six ulcer characteristics (size, number, duration, pain and site of ulcers) and a significant increase in the length of ulcer free periods at 6 months (p<0.01). Although further reduction had been noted in all ulcer characteristics at 12 months therapy, only the number of ulcers and the ulcer free periods continued to improve significantly for 12 months (p<0.05) while size, duration, pain and site reductions were not statistically significant compared with the 6 months score (p<0.4).

3.2.2 (Group 2) Colchicine tablet: Colchicine 500mcg OD

A total of 35 RAS patients who met the inclusion criteria were randomised in this group. They received one daily Colcihcine tablet 500mcg (OD) as monotherapy for 12 months, reviewed on 3 monthly intervals for a total of five visits. Nine patients (25.7%) (5 Females, 4 Males) withdrew treatment because of adverse events, work and family commitments and the data for these patients was not used. Therefore, this study was effectively conducted for 26 RAS subjects (age range 24-60 years, mean age 39 years). There were 12 Females (age range 24-60, mean age 38.2 years) and 14 Males (age range 28-59 years, mean age 39.7 years). In addition, 18 subjects were presented with Minor RAS and 8 with Major RAS.

Of the 26 subjects who were reviewed at 3 monthly intervals for 12 months, 2 (7.7%) participants had no recurrences during treatment period, 19 patients (73.1%) reported more than 20% improvement, while 5 subjects (19.2%) remained unchanged or became slightly worse. Four patients reported adverse events (11.4%) which necessitated termination of their therapy: 1 Female reported alopecia and 3 reported gastric pain and vomiting (1 Male and 2 Female). (Fig.3.11)

104

Demographic distribution of RAS patients in Colchicine group

side effect 4

withdraw 9

Finish 26

Randomise 35

0 10 20 30 40

Randomise Finish withdraw side effect RAS 35 26 9 4

Fig.3.11 Distribution of RAS patients in the Colchicine group, 35 randomised, 26 completed the 12 month follow up, 9 withdrew of which 4 reported side effects.

3.2.2.a Ulcer Severity Scores (USS) before and after systemic treatment with Colchicine:

The range of the USS for the 26 RAS patients before treatment with Colchicine tablets 500mcg OD was (20-48) with a mean of 32.5 ± 7 (SD). After 3 months therapy a significant reduction was seen in the USS to a mean of 22.2 ± 10.7 with a range of 0-50, and the mean percentage of improvement was 31.7% (p<0.001). There was no further significant improvement noted after 6 months, with a mean USS of 20.9 ± 9.9, giving a mean improvement of 35.7% (p<0.2). There was also no further significant improvement noted in the 9 month follow up, with a mean USS of 20.9 ± 10.8, giving a mean improvement of 35.7% (p<0.5). However, after 12 months therapy with Colchicine 500mcg OD there was a significant reduction in the USS to a mean of 17.7 ± 9.7 (SD) with a range of (0-32), giving a mean improvement of 45.5% (p<0.02). (Fig.3.12)

105

26 patients on Colchicine 500mcg OD followed for 12 months 40 31.7% 35.7% 35.7% 45.5% 35 30 * 25 * USS 20 15 10 5 0 baseline vit. 1 vit.2 vit. 3 vit. 4 USS 32.5 22.2 20.9 20.9 17.7

Fig.3.12 RAS patients (n=26) followed for 12 months on Colchicine tablet 0.5mg/day. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.05. ±SEM

3.2.2.b Comparison of Ulcer Severity Scores (USS) in Minor and Major RAS after systemic treatment with Colchicine:

There were 18 subjects with Minor RAS in this group, with a mean USS of 28.8 ± 4.4 (SD) and a range of 20-37. After 3 months treatment the mean USS was 20.2 ± 8.9 and the range was 0-31, giving a significant improvement of 29.9% (p<0.001). Although, no further statistically significant improvements were seen after this period compared with the previous visit, as in the 6 month follow up with a mean of 17.8 ± 9.6, (improvement of 38.2%) (p<0.1), 9 months with a mean of 17.1 ± 9.3, (improvement of 40.7%) (p<0.3) and in 12 month reviews with an improvement of 50.1%, range (0-30) and USS mean of 14.4 ± 8.7 (p<0.07), though there were significantly improved compared with the baseline visit (p<0.001). (Fig.3.13, Table 14)

106

18 Patients on Colchicine 500mcg OD followed for 12 months ( Minor RAS)

35 29.9% 38.2% 40.7 % 50.1% 30

25 *

20 USS 15

0 baseline vit.1 vit.2 vit.3 vit.4 USS 28.83 20.22 17.83 17.11 14.38

Fig.3.13 Minor RAS (n=18) patients followed for 12 months on systemic Colchicine 0.5mg/day. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.001. ±SEM

Analysis of the responses of individual patients showed considerable variations (Fig.3.13b). All of the patients with Minor RAS had an USS of higher than 20 at the beginning of the treatment period and 8 of the 18 patients had an USS of less than 20 after 3 months of therapy. Although 12 of the 18 subjects had an USS of less than 20 after 12 months therapy, it was not statistically significant (p<0.07). Some patients showed a steady decline in the ulcer severity over this time, while others showed a rapid improvement but then relapses. Little further improvement was found after 3 months, suggesting that this might be an optimal time to assess the efficacy of treatments. (Fig.3.13b)

107

pat 1 18 Patients on Colchicine 500mcg OD followed for 12 months pat 2 (Minor RAS) pat 3 pat 4 40 pat 5 pat 6 35 pat 7 30 pat 8 25 pat 9 pat 10 20 pat 11 15 pat 12 pat 13 10 pat 14 5 pat 15 0 pat 16 baseline vit.1 vit.2 vit.3 vit.4 pat 17 pat 18 Fig.3.13b The effect of Colchicine 0.5mg/day on individual Minor RAS patients (n=18) reviewed for 12 months, visit 1 (baseline), visit 2 (3 months), visit 3 (6 months), visit 4 (9 months) and visit 5 (12 months).

There were 8 subjects with Major RAS in this group (3 Females, 5 Males). The USS range before treatment was 35-48 with a mean of 40.6 ± 4.1 (SD). After 3 months therapy the mean USS was reduced to 26.5 ± 13.8, with a range of 0-50, giving a mean improvement of 34.8% (p<0.02). After 6 months there was a slight increase in the USS mean to 27.9 ± 6.9, with a decrease in the mean improvement of 31.4% (p<0.3). After 9 months there was a further increase in the USS mean to 29.5 ± 8.9, and a further decrease in the mean improvement to 27.4% (p<0.2). After 12 months of Colchicine 0.5mg OD as monotherapy the USS mean was 25 ± 8.1 (SD), with a range of 9-34, giving an increase in the mean improvement to 38.5%. However, this was not statistically significant compared with the previous visit (p<0.09) but still significantly less than baseline visit (p<0.001). (Fig.3.14, Table 14)

108

8 Patients on Colchicine 500mcg OD followed for 12 months (Major RAS) 45 34.8 % 31.4% 27.4% 40 38.5% 35 * 30 USS 25 20 15 10 5 0 baseline vit.1 vit.2 vit.3 vit.4 USS 40.62 26.5 27.87 29.5 25

Fig.3.14 Major RAS (n=8) patients followed for 12 months on systemic Colchicine 0.5mg/day. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.05.±SEM.

Analysis of the individual patients’ responses showed a considerable variation (Fig.3.14b). All of the patients with Major RAS had an USS of greater than 35 at the beginning of the treatment period and by 3 months, 6 out of 8 had an USS less than 30. Some patients showed a steady decline in the ulcer severity over this time, while others showed a rapid improvement but then relapses. One patient appeared to be recalcitrant to therapy as the USS mean score increased from 40 to 50 in the first 3 months. Little further improvement in the group was found after 3 months, suggesting that this might be a minimal period to assess the efficacy of treatments (Fig.3.14b, Table 14).

109

8 Patients on Colchicine 500mcg OD followed for 12 months (Major RAS)

60 pat 1 50 pat 2 40 pat 3 pat 4 30 pat 5 20 pat 6 pat 7 10 pat 8 0 baseline vit.1 vit.2 vit.3 vit.4

Fig.3.14b The effect of systemic Colchicine 0.5mg/day on individual Major RAS patients (n=8) reviewed for 12 months, visit 1 (baseline), visit 2 (3 months), visit 3 (6 months), visit 4 (9 months) and visit 5 (12 months).

Table.14 Summary of Ulcer Severity Scores before and after treatment with Colchicine 500mcg OD for 12 months in patients with Minor and Major RAS

Type of No. of USS USS USS USS % P value RAS patients range mean range mean improvements Compared before before after after with Rx Rx Rx Rx baseline Minor 18 20-37 28.8 0-30 14.4 50.1 % P<0.001 ± 4.4 ± 8.7 (SD) (SD)

Major 8 35-48 40.6 9-34 25 38. 5% P<0.001 ± 4.1 ± 8.1 (SD) (SD)

110

3.2.2.c Effect of Colchicine treatment on six individual characteristics of ulcers:

Each of the parameters for the six individual ulcer characteristics was examined before and after treatment to determine whether treatment with systemic Colchicine 0.5mg/day as monotherapy might have a differential effect. At the baseline visit the mean size score was 5.2 ± 2.3 (SD), which was reduced to 3.1 ± 2.1 after 6 months therapy and to 2.5 ± 1.1 after 12 months. The mean score of the number of ulcers at the baseline visit was 2.9 ± 0.9, reduced to 1.5 ± 0.9, after 6 months but it was slightly increased to 1.6 ± 1.5 after 12 months therapy. The mean score of the duration of ulcers was in constant reduction from the baseline visit mean score of 3.9 ± 2.2, down to 2.3 ± 1.7 after 6 months and to 1.8 ± 1.1 at the 12 months review. The mean score of the ulcer free periods at the baseline visit was 7.1 ± 2, which was decreased to 5.2 ± 3 after 6 months and to 3.7 ± 2.9 after 12 months therapy. At the baseline visit the mean score of the sites affected was 6.7 ± 2.2, which was reduced to 4.2 ± 2.8 at 6 months and 4.2 ± 3.1 at the 12 months review. At the baseline visit patients reported an average pain score of 6.8 ± 1.7, which was reduced after 6 months therapy to 4.7 ± 2.5 and to 3.8 ± 2.1 after 12 months treatment with Colchicine 0.5mg OD (SD). (Fig.3.15, Table 15)

Individual ulcer characteristics scores were found to be significantly decreased after 6 months treatment of Colchicine as monotherapy. Size of ulcers decreased by 40.4%, number of ulcers by 48.6%, duration of ulcers by 41.8%, pain by 30.6% and sites affected by 37.6%, as well as an increase in the length of ulcer free periods by 27.2% (p<0.001). After 12 months therapy, there were no further significant reductions reported in the size of the ulcer (50.9%) (p<0.2), number of ulcers (44.9%) (p<0.6), duration of the ulcer (53.6%) (p<0.2), pain (43.2%) (p<0.08) and site (37%) (p<0.8) compared with 6 months scores, though there were significantly decreased compared with the baseline visit (p<0.001). However, a further significant increase in the ulcer free periods by 48.4% was found, compared with scores at 6 months follow up (p<0.007). (Fig.3.15, Table 15)

111

26 patients on Colchicine 500mcg OD followed for 12 months 8 40.4% 48.6% 41.8% 7 27.2% 30.6% 37.6% 48.4% 6 53.6% 43.2% 37% 5 50.9% 4 44.9% 3 baseline 2 1 6 months 0 12 months size no duration ulcer free pain site baseline 5.15 2.92 3.88 7.07 6.76 6.65 6 months 3.07 1.5 2.26 5.15 4.69 4.15 12 months 2.53 1.61 1.8 3.65 3.84 4.19

Fig.3.15 Comparison of six ulcer characteristics scores (n=26 RAS patients) at baseline visit, 6 and 12 months reviews. Mean% improvement compared with the baseline visit, using paired t-test. ±SEM

Table.15 Summary of Ulcer characteristics scores before and after treatment with Colchicine 500mcg OD for 12 months

Ulcer Score (mean ) Score (mean) % improvements P value characteristics before Rx± (SD) after Rx ± (SD) (compared with baseline) Size 5.2 ± 2.3 2.5 ± 1.1 50.9 % P<0.001 No 2.9 ± 0.9 1.6 ± 1.5 44.9 % P<0.001 Duration 3.9 ± 2.2 1.8 ± 1.1 53.6 % P<0.001 Ulcer free 7.1 ± 2 3.7 ± 2.9 48.4 % P<0.001 Pain 6.8 ± 1.7 3.8 ± 2.1 43.2 % P<0.001 Site 6.7 ± 2.2 4.2 ± 3.1 37 % P<0.001

Overall clinical responses : 7.7% of patients had no recurrences during the treatment period, 19.2% did not show any change and 73.1% had a significant reduction in the USS using Colchicine 0.5mg OD as monotherapy for the duration of 12 months. 11.4% of patients developed an adverse event of alopecia, gastric pain and vomiting. Patients noted the benefits within the first 3 months of the treatment, with a maximum effect by 12 months (p<0.05).

112

The significant effect of Colchicine on Minor and Major RAS continued for up to 3 months only (p<0.05). Colchicine 0.5mg once a day had a significant effect on the reductions of size, number, duration, pain and site and increased the length of ulcer free periods in the first 6 months (p<0.01), but only the ulcer free periods continued their improvements significantly for 12 months (p<0.01).

3.2.3 (Group 3) Colchicine and Betnesol Group: Colchicine tablets 500mcg OD and Betnesol mouthwash QDS during ulcer attacks only

36 RAS patients were enrolled in this group who met the inclusion criteria. They received a combined therapy of Colchicine tablet 0.5mg OD and Betnesol mouthwash QDS during ulcer attacks. Five patients (13.9%) withdrew from treatment (4 Females, 1 Male) due to side effects and work or family commitments and the data for these subjects was not used. Therefore, this study was effectively conducted for 31 subjects, (age range 21-63 years, mean age 39.1 years). There were 15 Females (age range 26-55 years, mean age 39 years) and 16 Males (age range 21-63 years, mean age 39.2 years). 11 subjects were presented with Minor RAS while 20 were presented with Major RAS.

Of the 31 subjects who were reviewed at 3 monthly intervals for 12 months, one patient (3.2%) had no recurrences during the treatment period, 26 subjects (83.9%) were found to have a greater than 20% improvement in the USS, while 4 participants (12.9%) remained with no change or became slightly worse. Two patients (5.6%) experienced side effects (1 Male reported gastric pain and 1 Female reported peripheral neuropathy after 3 months of Colchicine plus Betnesol therapy), which was reversible after discontinuation of Colchicine. (Fig.3.16)

113

Demographic distribution of patients in Colchicine and Betnesol group

side effect 2

withdraw 5

Finish 31

Randomise 36

0 10 20 30 40

Randomise Finish withdraw side effect RAS 36 31 5 2

Fig.3.16 Distribution of RAS patients in Colchicine and Betnesol group, 36 randomised, 31 completed the 12 month follow ups and 5 withdrew of which 2 reported side effects.

3.2.3.a Ulcer Severity Scores (USS) before and after treatment with Colchicine and Betnesol:

The mean USS before treatment with Colchicine and Betnesol was 36.9 ± 6.5 (SD) (range 25-53) for the 31 subjects in this group. This was significantly decreased after 3 months therapy to a mean of 27.6 ± 7.7 and a range of 0- 41, representing a significant improvement of 25.2% (p<0.0001). After 6 months a further significant reduction was noticed to a mean of 23.1 ± 8.9, giving a mean improvement of 37.4% (p<0.001). After 9 months the USS mean was 20.2 ± 10.3, giving a significant improvement of 45.3% (p<0.02). Although there was a further small reduction in the mean USS after 12 months therapy to 19.2 ± 7.8 (SD), with a range of 0-34 and a mean improvement of 48%, this was not statistically significant (p<0.1) compared with the previous visit, though significantly less than the baseline score (p<0.001). (Fig.3.17)

114

31 patients on Colchicine and Betnesol followed for 12 months 45 40 25.2% 37.4% 45.3% 48% 35 * 30 * 25 * USS 20 15 10 5 0 baseline vit. 1 vit.2 vit. 3 vit. 4 USS 36.9 27.6 23.1 20.2 19.2

Fig.3.17 RAS patients (n=31) followed for 12 months on Colchicine and Betnesol. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. *p<0.05. ±SEM

3.2.3.b Ulcer Severity Scores (USS) in Minor and Major RAS after treatment with Colchicine and Betnesol:

There were 11 Minor RAS subjects in this group with an USS mean before treatment of 31 ± 3.3 (SD) and a range of 25-36. After 3 months treatment with Colchicine plus Betnesol, the mean USS decreased to 24.7 ± 3.8, representing a significant mean improvement of 20.3% (p<0.0005). After 6 months therapy the USS mean was 19.1 ± 8.5, giving a significant mean improvement of 38.4% (p<0.009). However, there were no further statistically significant improvements. At the 9 month follow up reviews the USS mean was 16.3 ± 10.6, giving a mean improvement of 47.5% (p<0.1). At the 12 month reviews the USS mean was slightly increased to 16.6 ± 7.2, representing a mean improvement of 46.4% (p<0.4) compared with the previous visit but remained significantly less than baseline visit (p<0.001). (Fig.3.18, Table 16)

115

11 Patients on Colchicine and Betnesol mouthwash (Minor RAS) 35 20.3% 38.4% 47.5% 46.4% 30 * 25 * USS 20 15 10 5 0 baseline 3 months 6 months 9 months 12 months USS 31 24.72 19.09 16.27 16.63

Fig.3.18 Minor RAS (n=11) patients followed for 12 months on Colchicine and Betnesol. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. * p<0.001. ±SEM

Analysis of the responses of individual patients showed considerable variation (Fig.3.18b). All of the patients with Minor RAS had an USS of greater than 25 at the beginning of the treatment period. 5 of the 11 patients had a USS of less than 20 after 6 months and two patients were recalcitrant to treatment. Some patients showed a steady decline in ulcer severity over this time, while others showed a rapid improvement but then relapses. Little improvement was found after 6 months, suggesting that this might be an optimal time to assess the efficacy of treatments. (Fig.3.18b)

116

11 Patients on Colchicine and Betnesol followed for 12 months (Minor RAS)

40 pat 1 35 pat 2 30 pat 3 pat 4 25 pat 5 20 pat 6 15 pat 7 10 pat 8 5 pat 9 0 pat 10 baseline vit. 1 vit.2 vit. 3 vit. 4 pat 11

Fig.3.18b The effect of Colchicine and Betnesol on individual Minor RAS patients (n=11) reviewed for 12 months, visit 1 (baseline), visit 2 (3 months), visit 3 (6 months), visit 4 (9 months) and visit 5 (12 months).

There were 20 participants with Major RAS in this group with a range of USS before treatment between (33-53) and a mean of 40.2 ± 5.5 (SD). After 3 months treatment with Colchicine 500mcg OD plus Betnesol mouthwash QDS a significant reduction of the mean USS to 29.3 ± 8.9 was seen (range 0-41), representing a mean improvement of 27.1% (p<0.0002). After 6 months the mean decreased to 25.3 ± 8.5, giving a mean improvement of 37% (p<0.001). After 9 months a further significant reduction was seen with a mean USS of 22.3 ± 9.8, giving a 44.5% improvement (p<0.03). Although after 12 months therapy there was a further reduction in the USS mean to 20.6 with an improvement of 48.7% compared with the baseline visit (p<0.001), this was not statistically significant compared with the 9 months score (p<0.09). (Fig.3.19, Table 16)

117

20 Patients on Colchicine and Betnesol mouthwash (Major RAS)

45 40 27.1% 37% 44.5% 48.7% 35 * * 30 * USS 25 20 15 10 5 0 baseline 3 months 6 months 9 months 12 months USS 40.15 29.25 25.3 22.3 20.6

Fig.3.19 Major RAS (n=20) patients followed for 12 months on Colchicine and Betnesol. Figures above columns show the USS mean % improvement compared with the baseline and significances compared with the previous visit using paired t-test. * p<0.05. ±SEM

Individual patients’ responses showed a considerable variation (Fig.3.19b). All of the patients with Major RAS had an USS of greater than 30 at the beginning of the treatment period and by 9 months, 17 of the 20 patients responded to treatment, two patient appeared to be recalcitrant to therapy, and in one subject ulcers virtually disappeared after 3 months therapy with Colchicine plus Betnesol. Patients showed a steady decline in ulcer severity over this time, while others showed a rapid improvement, but then relapses. Little improvement in the group was found after 9 months, suggesting that this might be an optimal period to assess the efficacy of treatments (Fig.3.19b).

118

pat 1 20 Patients on Colchicine and Betnesol followed for 12 months pat 2 ( Major RAS) pat 3 pat 4 pat 5 60 pat 6 pat 7 50 pat 8 pat 9 40 pat 10 pat 11 30 pat 12 pat 13 20 pat 14 pat 15 pat 16 10 pat 17 pat 18 0 pat 19 1 2 3 4 5 pat 20

Fig.3.19b The effect of Colchicine and Betnesol on individual Major RAS patients (n=20) reviewed for 12 months, visit 1 (baseline), visit 2 (3 months), visit 3 (6 months), visit 4 (9 months) and visit 5 (12 months).

Table.16 Summary of Ulcer Severity Score (USS) before and after treatment with Colchicine 500mcg OD and Betnesol mouthwash QDS for 12 months in patients with Minor and Major RAS

Type of No. of USS USS USS USS % P value RAS patients range mean range Mean improvement compared before before after after with Rx Rx Rx Rx baseline

Minor 11 25-36 31 8-33 16.6 46.4 % P<0.001 ± 3.3 ± 7.2 (SD) (SD)

Major 20 33-53 40.2 0-34 20.6 48.7 % P<0.001 ± 5.5 ± 7.9 (SD) (SD)

119

3.2.3.c Effect of Colchicine and Betnesol treatment on six individual characteristics of ulcers:

Each of the six ulcer characteristics (size, number, duration, ulcer free periods, pain and site), was examined before and after treatment to determine whether treatment with Colchicine plus Betnesol mouthwash might have a differential effect. At the baseline visit the mean score for size was 5.7 ± 2.5 (SD), which was reduced to 3.1 ± 1.5 after 6 months and to 2.6 ± 1.1 after 12 months therapy. At the baseline visit the mean score for number of ulcers was 3.8 ± 1.1, which was reduced to 1.9 ± 1 (6 months) and to 1.6 ± 0.9 after 12 months. The mean score for duration of ulcers was in constant reduction, from 4.3 ± 2.2 at the baseline visit to 2.2 ± 1.6 at the 6 months visit and finally to 2 ± 1.5 at the 12 months review. At the baseline visit the mean score of the ulcer free periods was 7.9 ± 2, which was significantly decreased after 6 months to 6.1 ± 2.8 and after 12 months to 5 ± 3.1. At the baseline visit the mean score of the sites affected was 7.8 ± 2.3, which was decreased after 6 months to 4.9 ± 2.7 and to 4.1 ± 2.6 at 12 months visit. At the baseline visit patients gave an average pain score of 7.1 ± 1.9, which was reduced to 4.6 ± 2.2 at 6 months with a further significant reduction to 3.7 ± 1.6 (SD) after 12 months treatment.

Scores for individual ulcer characteristics were found to be significantly decreased after 6 months treatment with Colchicine plus Betnesol mouthwash. The main reductions were: the size of the ulcers by 45.2%, number of the ulcers by 49.4%, duration of the ulcers by 47%, pain by 35.3%, sites affected by 37.4% as well as an increase in the length of ulcer free periods by 22.6% (p<0.001). However, further significant reductions were noticed after 12 months in the size of ulcers by 54.2% (p<0.03), in pain score by 47.5% (p<0.008) and an increase in the length of ulcer free periods by 36.1% (p<0.04). The reductions in number of the ulcers by 57.6% (p<0.09), duration of the ulcers by 53.7% (p<0.06) and sites affected by 47.1% (p<0.1) were not statistically significant compared with the 6 months scores but significantly decreased compared with the baseline scores (p<0.001). (Fig.3.20, Table 17) 120

31 patients on Colchicine and Betnesol folllowed for 12 months 9 45.2% 49.4% 47% 22.6% 35.3% 8 37.4% 7 36.1% 47.5% 47.1% 6 baseline 5 54.2% 57.6% 53.7% 6 months 4 3 12 months 2 1 0 size no duration ulcer free pain site baseline 5.77 3.87 4.32 7.96 7.12 7.87 6 months 3.16 1.96 2.29 6.16 4.61 4.93 12 months 2.64 1.64 2 5.09 3.74 4.16

Fig.3.20 Comparison of six ulcer characteristics scores (n=31 RAS patients) at baseline visit, 6 and 12 months reviews. Mean % improvement compared with the baseline visit using paired t-test. ±SEM

Table.17 Summary of ulcer characteristics scores before and after treatment with Colchicine 500mcg OD and Betnesol mouthwash QDS for 12 months

Ulcer Score (mean) Score (mean) % improvements P value characteristics before Rx ± after Rx ± (SD) (compared with (SD) baseline) Size 5.8 ± 2.5 2.6 ± 1.1 54.2% P<0.001 Number 3.9 ± 1.1 1.6 ± 0.9 57.6% P<0.001 Duration 4.3 ± 2.2 2 ± 1.5 53.7% P<0.001 Ulcer free 8 ± 2 5.1 ± 3.1 36.1% P<0.001 Pai n 7.1 ± 1.9 3.7 ± 1.6 47.5% P<0.001 Site 7.9 ± 2.3 4.2 ± 2.6 47.1% P<0.001

Overall clinical responses: 3.2% of patients had no recurrences during the treatment period and 83.9% showed a reduction in the USS with Colchicine 0.5mg OD and Betnesol mouthwash QDS during ulcer attacks. 5.6% of the participants reported side effects of peripheral neuropathy and gastric pain after 3 months therapy. Although Major RAS patients noted the benefits of the combined therapies in the first 3 months (p<0.001), the USS continued to

121 improve with the maximum significant effects by 9 months (p<0.02). The significant effects on Minor RAS continued up to 6 months (p<0.009) but the further improvement to 9 months was not statistically significant, probably due to two recalcitrant patient. Otherwise the profiles of Major and Minor RAS were very similar. RAS patients on the combined therapy had a significant reduction in all ulcer characteristics (size, number, duration, pain and site of the ulcer) and an increase in the length of ulcer free periods at 6 months (p<0.001). Although further reductions had been noted in all ulcer characteristics after 12 months therapy, only size of the ulcers, pain score and ulcer free periods reached statistical significances (p<0.05).

3.3 Comparison of Colchicine 0.5mg/day with 1mg/day in 10 Major RAS patients:

In this study a total of 10 patients with Major RAS were recalcitrant to treatment with Colchicine 0.5mg a day for 12 months. These patients were followed for a further 6 months in which they received Colchicine tablets 1mg a day.

The mean of the USS for this group before treatment with Colchicine 1mg OD was 26 ± 5.7 (SD) with a range of 18-34. After 6 months therapy with Colchicine 1mg/day, a significant reduction was seen in the USS to a mean of 18.6 ± 7.2, with a range of 10-34, representing a mean improvement of 28.5% (p<0.001) compared with Colchicine 0.5mg OD at 12 month follow ups (Fig.3.22). The mean percentage of improvement for this group was 51.3% compared with the baseline visit (p<0.0001) (Fig.3.21).

122

10 major RAS patients on Colchicine 0.5mg followed for 12 months then 1mg for 6 months (% improvement in USS) 50 45 63.4% 68.2% 66.7% 44.4% 40 0% 62.2% 35 71.4% 32.4% 41.7% 54.3% 30 25 20 15 basseline 10 12 months 5 0 18 nonths pat pat 1 pat 2 pat 3 pat 4 pat 5 pat 6 pat 7 pat 8 pat 9 10 basseline 35 41 34 44 36 33 42 37 45 35 12 months 18 27 32 20 22 34 24 24 34 25 18 nonths 10 15 23 14 21 34 14 14 25 16

Fig.3.21 Major RAS patients (n=10) followed for 12 months who were on Colchicine 0.5mg/day then increased to 1mg/day for 6 months. Figures above columns show the USS mean % improvement compared with the baseline.

Comparison of Colchicine 0.5mg/day with 1mg/day for 6 months in 10 Major RAS patients (% improvement in USS) 40 0% 26.5% 35 28.1% 30 44% 36% 25 4.5% 41% 41.7% 30% 20 44.4% 15 10 5 0 pat 1 pat 2 pat 3 pat 4 pat 5 pat 6 pat 7 pat 8 pat 9 pat 10

Fig.3.22 Major RAS patients (n=10) followed for 6 months who were on Colchicine 0.5mg/day for 12 months then increased to 1mg/day for 6 months. Figures above columns show the USS mean % improvement compared with the 12 month reviews on Colchicine 0.5mg/day.

123

Of the 10 patients in this group who received Colchicine 1mg a day for 6 months, there were 2 subjects who were still recalcitrant to treatment (patient 5 and 6 in Fig.3.23), of whom one patient reported the adverse event of gastrointestinal disturbances which necessitated termination of Colchicine.

10 Major RAS patients on Colchicine 0.5mg/day followed for 12 months then on 1mg/day for 6 months (USS) 50 45 pat 1 40 pat 2 35 30 pat 3 25 pat 4 20 pat 5 15 10 pat 6 5 pat 7 0 pat 8 baseline 3 months 6 months 9 months 12 months 18 months

Fig.3.23 USS of the individual Major RAS patients (n=10) who were on Colchicine 0.5mg/day for 12 months (baseline, 3 months, 6 months, 9 months, 12 months) and then on Colchicine 1mg/day for 6 months (follow up point of the latter dose in the graph is 18 months).

124

CHAPTER FOUR RESULTS Cytokines profile of Recurrent Aphthous Stomatitis (RAS) and the effects of treatments on serum and salivary cytokines

4.1 Introduction:

Nine cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g) were measured in serum and salivary samples of 72 RAS patients compared with 34 healthy controls using Luminex Multianalyte Profiling System (MAP). Five salivary cytokines (IL-8, IL-6, IL-10, TNF-a and IFN-g) were raised in RAS patients compared with healthy controls (p<0.05), while the mean concentrations of four salivary cytokines (IL-2, IL4, IL5 and IL-17) were not significantly different from controls.

In serum, IL-8 was the only cytokine which was detected significantly raised in the RAS subjects compared with healthy controls, though in a level that was much lower than salivary concentrations.

4.2 Cytokines in Healthy controls and RAS patients:

Salivary and serum IL-8 is raised in RAS:

Salivary level of IL-8 in the 72 RAS patients, with a mean of 4027.5 ± 2051 pg/ml (SD) and a range of 703.3-10288.3 pg/ml, which was significantly higher compared with 34 healthy controls with a mean of 2476.2 ± 1220.4 and a range of 333-5447 pg/ml (p<0.0001). (Fig.4.1)

125

Controls and RAS patients IL-8 (saliva) 4500 * 4000 3500 3000 2500 2000 pg/ml 1500 1000 500 0 Controls(34) before Rx(72) pg/ml 2476.2 4027.5

Fig.4.1 Comparison of IL-8 concentrations in saliva of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test *(p<0.0001). Mean ± SEM

Serum level of IL-8 in RAS subjects, with a mean of 98.9 ± 214 (SD) and a range of 10.5-1209.5 pg/ml, which was significantly higher compared with healthy controls with a mean of 47.8 ± 32.7 and a range of 17.3-183.5 pg/ml (p<0.04). (Fig.4.2)

Control and RAS patients IL-8 (serum)

140 * 120 100 80 60 pg/ml 40 20 0 Controls(34) before Rx(72) pg/ml 47.8 98.9

Fig.4.2 Comparison of IL-8 concentrations in serum of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test *(p<0.04). Mean ± SEM

126

Salivary but not serum IL-6 is raised in RAS:

Salivary level of IL-6 in RAS patients, with a mean of 74 ± 149.6 (SD) and a range of 2.3-1078 pg/ml, was significantly higher than in healthy controls with a mean of 7.4 ± 9.5 and a range of 1.5-59 pg/ml (p<0.0001). (Fig.4.3)

Controls and RAS patients IL-6 (saliva) 100 * 90 80 70 60 50 40 pg/ml 30 20 10 0 Controls(34) before Rx(72) pg/ml 7.4 74

Fig.4.3 Comparison of IL-6 concentrations in saliva of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test *(p<0.0001). Mean ± SEM

Serum level of IL-6 in RAS subjects, with a mean of 1.6 ± 0.6 and a range of 0.1-4 pg/ml, was not significantly different from healthy controls with a mean of 1.7 ± 1.3 and a range of 0.5-8.5 pg/ml (p<0.3). (Fig.4.4)

127

Controls and RAS patients IL-6 (serum) 2 1.8 1.6 1.4 1.2 1 0.8 pg/ml 0.6 0.4 0.2 0 Controls(34) before Rx(72) pg/ml 1.7 1.6

Fig.4.4 Comparison of IL-6 concentrations in serum of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test (p<0.3). Mean ± SEM

Salivary but not serum IL-10 is raised in RAS:

Salivary level of IL-10 was significantly raised in RAS patients with a mean of 8.3 ± 8.8 (SD) and a range of 1-37.5 pg/ml, compared with healthy controls with a mean of 4.2 ± 3.7 and a range of 1-16.5 (p<0.0004). (Fig.4.5)

Controls and RAS patients IL-10 (saliva) 10 * 9 8 7 6 5 4 pg/ml 3 2 1 0 Controls(34) before Rx(72) pg/ml 4.2 8.3

Fig.4.5 Comparison of IL-10 concentrations in saliva of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test *(p<0.0004). Mean ± SEM

128

Serum level of IL-10 was not significantly different between RAS subjects with a mean of 3.2 ± 4.6 and controls with a mean of 2.8 ± 1 (p<0.2). (Fig.4.6)

Controls and RAS patients IL-10 (serum) 4 3.5 3 2.5 2 1.5 pg/ml 1 0.5 0 Controls(34) before Rx(72) pg/ml 2.8 3.2

Fig.4.6 Comparison of IL-10 concentrations in serum of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test (p<0.2). Mean ± SEM

Salivary but not serum TNF-a is raised in RAS:

Salivary level of TNF-a was found significantly higher in RAS patients, with a mean of 13.6 ± 19.5 (SD) and a range of 1.5-76.3 pg/ml, compared with healthy controls with a mean of 4.6 ± 2.6 and a range of 1.8-12.8 pg/ml (p<0.0001). (Fig.4.7)

129

Controls and RAS patients TNF-a (saliva) 18 * 16 14 12 10 8 pg/ml 6 4 2 0 Controls(34) before Rx(72) pg/ml 4.6 13.6

Fig.4.7 Comparison of TNF-a concentrations in saliva of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test *(p<0.0001). Mean ± SEM

The serum level of TNF-a was not significantly different in RAS subjects, with a mean of 3.8 ±1.3 and a range of 1.3-8.8 pg/ml, compared with healthy controls with a mean of 3.9 ± 1.4 and a range of 2-7 pg/ml (p<0.3). (Fig.4.8)

Controls and RAS patients TNF-a (serum) 4.5 4 3.5 3 2.5 2 pg/ml 1.5 1 0.5 0 Controls(34) before Rx(72) pg/ml 3.9 3.8 Fig.4.8 Comparison of TNF-a concentrations in serum of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test (p<0.3). Mean ± SEM

130

Salivary but not serum IFN-g is raised in RAS:

Salivary concentration of IFN-g was significantly higher in RAS subjects, with a mean of 11 ± 32.7 and a range of 1.5-225.7 pg/ml, compared with healthy controls with a mean of 3.3 ± 0.7 and a range of 1.5-5 pg/ml (p<0.02). (Fig.4.9)

Controls and RAS patients IFN-g (saliva) 16 * 14 12 10 8 6 pg/ml 4 2 0 Controls(34) before Rx(72) pg/ml 3.3 11

Fig.4.9 Comparison of IFN-g concentrations in saliva of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test *(p<0.02). Mean ± SEM

Serum level of IFN-g was not significantly different in RAS subjects, with a mean of 2.8 ± 0.8 and a range of 0.8-6 pg/ml, compared with healthy controls with a mean of 3 ± 1.6 and a range of 1-11.5 pg/ml (p<0.2). (Fig.4.10)

131

Controls and RAS patients IFN-g (serum) 3.5 3 2.5 2 1.5 pg/ml 1 0.5 0 Controls(34) before Rx(72) pg/ml 3 2.8

Fig.4.10 Comparison of IFN-g concentrations in serum of 72 RAS patients and 34 healthy controls assayed by Luminex using unpaired t-test (p<0.2). Mean ± SEM

In conclusion, five salivary pro-inflammatory cytokines (IL-6, IL-8, IL-10, TNF- a and IFN-g) were found to be significantly increased in RAS patients compared with healthy controls, while in serum IL-8 was the only cytokine which was raised in RAS subjects compared with controls. IL-2, IL-4, IL-5 and IL-17 were not found to be different from controls either in saliva or in serum. Since these salivary and serum cytokines were raised in the RAS patients, we then raised the question, whether this was directly related to the presence of ulcers, or whether this was inherent to subjects with RAS irrespective of the presence of ulcers.

4.3 Cytokines in relation to the presence of ulcers:

In order to determine whether the presence of ulcers was responsible for the raised value of cytokines, 72 RAS patients were divided further into 41 subjects who had ulcers at the time of collection of the samples and 31 subjects who did not have ulcers and results were compared with 34 healthy controls.

132

Salivary IL-8 levels are independent of the presence of ulcers:

The mean salivary IL-8 concentration in the 41 ulcer present subjects was 4259.4 ± 2014.2 (SD), with a range of 703.3-10288.3 pg/ml. This was not significantly different from the 31 non-ulcer present subjects who had a mean of 3624.3 ± 2086.5, with a range of 842-9534.3 pg/ml (p<0.1). Salivary concentration of IL-8 was significantly raised in both the ulcer present group (41) (p<0.0001) and in the non-ulcer group (31) (p<0.002) compared with controls (34). (Fig.4.11)

Serum IL-8 levels are independent of the presence of ulcers:

The mean serum IL-8 concentration of the ulcer present subjects was 97.1 ± 212.6, with a range of 10.5-1209.5 pg/ml. This was not significantly different (p<0.4) from the non-ulcer present subjects who had a mean of 108.4 ± 219.4, with a range of 13.7-1041 pg/ml). There were no statistically significant differences between both RAS groups and healthy controls (p<0.09). (Fig.4.12)

With and without ulcer in RAS patients IL-8 (saliva) 5000 * 4500 * 4000 3500 3000 2500 2000 pg/ml 1500 1000 500 0 Controls(34) No ulcer(31) ulcer(41) pg/ml 2476.2 3624.3 4259.4

Fig.4.11 Comparison of salivary IL-8 levels in RAS patients with (n=41) or without (n=31) ulcers. Both groups were significantly greater than controls but not significantly different from each other.*(p<0.001) compared with controls using unpaired t-test. Mean ± SEM

133

With and without ulcer in RAS patients IL-8 (serum) 160 140 120 100 80 60 pg/ml 40 20 0 Controls(34) No ulcer(31) ulcer(41) pg/ml 47.8 108.4 97.1

Fig.4.12 Comparison of IL-8 levels in serum of RAS patients with (n=41) or without (n=31) ulcers. There were no significant differences between the groups (p<0.09) using unpaired t-test. Mean ± SEM

Salivary IL-6 levels are further raised in the presence of ulcers:

The mean salivary IL-6 concentration in the ulcer present subjects was 96.2 ± 190.1 (SD), with a range of 2.5-1078 pg/ml. This was significantly raised compared with the non-ulcer subjects who had a mean of 40.8 ± 53.5, with a range of 2.3-184.3 pg/ml (p<0.05). The mean concentrations of IL-6 in both groups were significantly higher than controls (p<0.001). (Fig.4.13)

With and without ulcer in RAS patients IL-6 (saliva) 140 * 120 100 80 x 60 * pg/ml 40 20 0 Controls(34) No ulcer(31) ulcer(41) pg/ml 7.4 40.8 96.2

134

Fig.4.13 Comparison of salivary IL-6 levels in RAS patients with (n=41) or without (n=31) ulcers. X= Salivary IL-6 significantly higher in ulcer present group than non-ulcer (p<0.05). Both groups significantly greater than controls *(p<0.001) using unpaired t-test. Mean ± SEM

Serum IL-6 concentrations were detected at very low levels in both RAS patient groups and in the control group. The ulcer present subjects had a mean of 1.5 ± 0.5 (SD), with a range of 0.5-3 pg/ml, compared with the non- ulcer subjects who had a mean of 1.7 ± 0.7, with a range of 4-0.8 pg/ml. The healthy controls had a mean of 1.7 ± 1.3 with a range of 0.5-8.5 pg/ml. No differences were found between all three groups (p<0.1). (Fig.4.14)

With and without ulcer in RAS patients IL-6 (serum) 2 1.8 1.6 1.4 1.2 1 0.8 pg/ml 0.6 0.4 0.2 0 Controls(34) No ulcer(31) ulcer(41) pg/ml 1.7 1.7 1.5

Fig.4.14 Comparison of IL-6 levels in serum of RAS patients with (n=41) or without (n=31) ulcers. There were no significant differences between the groups (p<0.1) using unpaired t-test. Mean ± SEM

Salivary IL-10 levels were not related to the presence of ulcers:

The salivary concentrations of IL-10 were significantly higher in the ulcer present group compared with controls (p<0.001), but were not statistically significant between the non-ulcer and control groups (p<0.07). The mean salivary IL-10 concentration in the ulcer present group was 9.4 ± 8.6 (SD), with a range of 1.5-37.5 pg/ml. This was not significantly different from the non-ulcer group who had a mean of 6.8 ± 9 (p<0.1), with a range of 1-37 pg/ml (Fig.4.15). 135

With and without ulcer in RAS patients IL-10 (saliva) 12 * 10

6 pg/ml 4

0 Controls(34) No ulcer(31) ulcer(41) pg/ml 4.2 6.8 9.4

Fig.4.15 Comparison of salivary IL-10 levels in RAS patients with (n=41) or without (n=31) ulcers. The groups were not significantly different from each other (p<0.1) but the ulcer present group was significantly higher than controls *(p<0.001) using unpaired t-test. Mean ± SEM

Serum IL-10 concentrations were present at low levels in both patients and controls and no differences between the three groups were apparent. (Fig.4.16)

With and without ulcer in RAS patients IL-10 (serum) 6

3 pg/ml 2

0 Controls(34) No ulcer(31) ulcer(41) pg/ml 2.8 3.7 2.8

Fig.4.16 Comparison of IL-10 levels in serum of RAS patients with (n=41) or without (n=31) ulcers. There were no significant differences between the groups (p<0.2) using unpaired t-test. Mean ± SEM

136

Salivary TNF-a levels were not related to the presence of ulcers:

The mean salivary TNF-a concentration in the ulcer present group was 15.5 ± 24.1 (SD), with a range of 1.8-141.5 pg/ml. This was not significantly different from the non-ulcer group who had a mean of 11 ± 10.5 (p<0.1), with a range of 1.5-44 pg/ml. The mean level of TNF-a was significantly higher in both groups compared with controls (p<0.002). (Fig.4.17)

With and without ulcer in RAS patients TNF-a (saliva) 25 * 20

15 *

10 pg/ml

0 Controls(34) No ulcer(31) ulcer(41) pg/ml 4.6 11 15.5

Fig.4.17 Comparison of salivary TNF-a levels in RAS patients with (n=41) or without (n=31) ulcers. Both groups were significantly greater than controls but not significantly different from each other. *(p<0.002) compared with controls using unpaired t-test. Mean ± SEM

Serum concentrations of TNF-a were detected at low levels in both RAS patients and controls and no significant differences were found (p<0.3). (Fig.4.18)

137

With and without ulcer in RAS patients TNF-a (serum) 4.5 4 3.5 3 2.5 2 pg/ml 1.5 1 0.5 0 Controls(34) No ulcer(31) ulcer(41) pg/ml 3.9 3.8 3.8

Fig.4.18 Comparison of TNF-a levels in serum of RAS patients with (n=41) or without (n=31) ulcers. There were no significant differences between the groups (p<0.3) using unpaired t-test. Mean ± SEM

Salivary IFN-g levels are raised in the presence of ulcers:

The mean salivary IFN-g concentration in the ulcer present subjects was 15.7 ± 42.6 (SD), with a range of 2-225.7 pg/ml. This was significantly raised compared with the non-ulcer group who had a mean of 4.1 ± 3.3 (p<0.04), with a range of 1.5-20.5 pg/ml. Salivary levels of IFN-g were significantly higher in the ulcer present group compared with controls (p<0.05). (Fig.4.20)

With and without ulcer in RAS patients IFN-g (saliva) 25 * 20 x 15

10 pg/ml

0 Controls(34) No ulcer(31) ulcer(41) pg/ml 3.3 4.1 15.7

138

Fig.4.20 Comparison of salivary IFN-g levels in RAS patients with (n=41) or without (n=31) ulcers. X= Salivary IFN-g significantly higher in ulcer present group than non-ulcer (p<0.04) and only ulcer group significantly greater than controls *(p<0.05) using unpaired t-test. Mean ± SEM

Serum IFN-g concentrations were detected at very low levels in both patients and controls and no differences were found between all three groups (p<0.2). (Fig.4.21)

With and without ulcer in RAS patients IFN-g (serum) 3.5 3 2.5 2 1.5 pg/ml 1 0.5 0 Controls(34) No ulcer(31) ulcer(41) pg/ml 3 2.8 2.9

Fig.4.21 Comparison of IFN-g levels in serum of RAS patients with (n=41) or without (n=31) ulcers. There were no significant differences between the groups (p<0.2) using unpaired t-test. Mean ± SEM

In conclusion, two salivary pro-inflammatory cytokines (IL-6 and IFN-g) were found to be significantly raised in RAS patients (p<0.05) in the presence of ulcers when compared with non-ulcers. This suggests that the levels of IL-6 and IFN-g in saliva are directly related to the inflammatory process.

In serum, no cytokines were found to be raised in the presence of ulcers compared with non-ulcers.

Since some of the cytokines in saliva were raised in the presence of ulcers (IL-6 and INF-g), a question was raised about whether this might be related to the severity of ulcers. Thus, a comparison of Major and Minor RAS was made. 139

4.4 Comparison of cytokines in Minor and Major RAS:

In order to determine whether the severity of RAS was responsible for the raised values of cytokines, further division of the 72 RAS patients was carried out, forming subgroups according to the disease types: (37) Minor and (35) Major compared with (34) healthy controls.

Salivary IL-8 levels were not different in Major and Minor RAS:

The mean salivary IL-8 concentration in the 37 Minor RAS patients was 3864.7 ± 1898.7 (SD), with a range of 842-8558 pg/ml. This was not significantly different from the 35 Major RAS subjects who had a mean of 4327 ± 2200.6, with a range of 703.3-10288.3 pg/ml (p<0.1). Salivary concentrations of IL-8 were significantly raised in both groups of Minor and Major RAS compared with healthy controls (p<0.001). (Fig.4.22)

Comparison of cytokines in Minor and Major RAS IL-8 (saliva) 5000 * 4500 * 4000 3500 3000 2500 2000 pg/ml 1500 1000 500 0 Controls(34) minor (37) major(35) pg/ml 2476.2 3864.7 4327

Fig.4.22 Comparison of salivary IL-8 levels in RAS patients with Minor (n=37) or Major (n=35). Both groups were significantly greater than controls but not significantly different from each other. *(p<0.001) compared with controls using unpaired t-test. Mean ± SEM

The mean serum IL-8 concentration in the Minor group was 113.5 ± 163.4, with a range of 10.8-734 pg/ml. This was not significantly different from the Major group which had a mean of 148.4 ± 259.1, with a range of 10.5-1209.5

140 pg/ml (p<0.3). Neither were statistically significant compared with healthy controls (p<0.09). (Fig.4.23)

Comparison of cytokines in Minor and Major RAS IL-8 (serum) 250

200

150

100 pg/ml

0 Controls(34) minor (37) major(35) pg/ml 47.8 113.5 148.4

Fig.4.23 Comparison of IL-8 levels in serum of RAS patients with Minor (n=37) or Major (n=35). There were no significant differences between the groups (p<0.3) using unpaired t-test. Mean ±SEM

Salivary IL-6 levels are greater in Major than Minor RAS:

The mean salivary IL-6 concentration in the Minor RAS group was 40.6 ± 50.4 (SD), with a range of 2.3-184.3 pg/ml. This was significantly different from the Major group, which had a mean of 87.5 ± 203.8, with a range of 4-1078 pg/ml (p<0.04). Salivary concentrations of IL-6 in both groups were significantly raised compared with healthy controls (p<0.001). (Fig.4.24)

141

Comparison of cytokines in Minor and Major RAS IL-6 (saliva) 140 * 120 100 x 80 60 * pg/ml 40 20 0 Controls(34) minor (37) major(35) pg/ml 7.4 40.6 87.5

Fig.4.24 Comparison of salivary IL-6 levels in RAS patients with Minor (n=37) or Major (n=35). X= Salivary IL-6 significantly higher in Major RAS group than Minor (p<0.04) and both groups significantly greater than controls *(p<0.001) using unpaired t-test. Mean ± SEM

Serum levels of IL-6 were not significantly different between the groups (p<0.2). (Fig.4.25)

Cytokines in Minor and Major RAS IL-6 (serum) 2 1.8 1.6 1.4 1.2 1 0.8 pg/ml 0.6 0.4 0.2 0 Controls(34) minor (37) major(35) pg/ml 1.7 1.6 1.6

Fig.4.25 Comparison of IL-6 levels in serum of RAS patients with Minor (n=37) or Major (n=35). There were no significant differences between the groups (p<0.2) using unpaired t-test. Mean ± SEM

142

Salivary IL-10 levels were not different in Major and Minor RAS :

The mean salivary IL-10 concentration in the Minor RAS group was 8 ± 9.7 (SD), with a range of 1-37.5 pg/ml. This was not significantly different from the Major group which had a mean of 7.4 ± 7.8, with a range of 1.5-32 pg/ml (p<0.3). Salivary levels of IL-10 in both groups were significantly raised compared with controls (p<0.02). (Fig.4.26)

Comparison of cytokines in Minor and Major RAS IL-10 (saliva) 12 * 10 * 8

6 pg/ml 4

0 Controls(34) minor (37) major(35) pg/ml 4.2 8 7.4

Fig.4.26 Comparison of salivary IL-10 levels in RAS patients with Minor (n=37) or Major (n=35). Both groups were significantly greater than controls but not significantly different from each other. *(p<0.02) compared with controls using unpaired t-test. Mean ± SEM

Serum concentrations of IL-10 were present at low levels in both patients and controls and no differences between the three groups were apparent. (Fig.4.27)

143

Comparison of cytokines in Minor and Major RAS IL-10 (serum) 5 4.5 4 3.5 3 2.5 2 pg/ml 1.5 1 0.5 0 Controls(34) minor (37) major(35) pg/ml 2.8 3.5 2.6

Fig.4.27 Comparison of IL-10 levels in serum of RAS patients with Minor (n=37) or Major (n=35). There were no significant differences between the groups (p<0.1) using unpaired t-test. Mean ± SEM

Salivary TNF-a levels were not different in Major and Minor RAS:

The mean salivary TNF-a concentration in the Minor RAS group was 9.8 ± 9.2 (SD), with a range of 1.5-44 pg/ml. This was not significantly different compared with the Major RAS group which had a mean of 16.3 ± 26 and a range of 1.8-141.5 pg/ml (p<0.07). Salivary concentrations of TNF-a in both groups were significantly higher compared with controls (p<0.001). (Fig.4.28)

Comparison of cytokines in Minor and Major RAS TNF-a (saliva) 25 * 20

15 * 10 pg/ml

0 Controls(34) minor (37) major(35) pg/ml 4.6 9.8 16.3

144

Fig.4.28 Comparison of salivary TNF-a levels in RAS patients with Minor (n=37) or Major (n=35). Both groups were significantly greater than controls but not significantly different from each other. * (p<0.001) compared with controls using unpaired t-test. Mean ± SEM

Serum levels of TNF-a were not significantly different between the three groups (p<0.4). (Fig.4.29)

Comparison of cytokines in Minor and Major RAS TNF-a (serum) 5 4.5 4 3.5 3 2.5 2 pg/ml 1.5 1 0.5 0 Controls(34) minor (37) major(35) pg/ml 3.9 3.9 4.2

Fig.4.29 Comparison of TNF-a levels in serum of RAS patients with Minor (n=37) or Major (n=35). There were no significant differences between the groups (p<0.4) using unpaired t-test. Mean ± SEM

Salivary IFN-g levels were raised in Major RAS:

The mean salivary IFN-g concentration in the Major RAS group was 13.4 ± 45.9 (SD), with a range of 2-225.7 pg/ml. This was significantly higher than the Minor group which had a mean of 4 ± 3.1 and a range of 1.5-20.5 pg/ml (p<0.04). Salivary concentration of IFN-g was significantly raised in the Major group compared with controls (p<0.03) whereas no significant differences were noticed in the Minor RAS group compared to controls (p<0.08). (Fig.4.30)

145

Comparison of cytokines in Minor and Major RAS IFN-g (saliva) 25 * 20

15 x 10 pg/ml

0 Controls(34) minor (37) major(35) pg/ml 3.3 4 13.4

Fig.4.30 Comparison of salivary IFN-g levels in RAS patients with Minor (n=37) or Major (n=35). X=Salivary IFN-g significantly higher in Major group than Minor (p<0.04) but only Major group significantly greater than controls *(p<0.05) using unpaired t-test. Mean ± SE

Serum concentrations of IFN-g were not significantly different between all three groups (p<0.4). (Fig.4.31)

Comparison of cytokines in Minor and Major RAS IFN-g (serum) 3.5 3 2.5 2 1.5 pg/ml 1 0.5 0 Controls(34) minor (37) major(35) pg/ml 3 2.8 2.8

Fig.4.31 Comparison of IFN-g levels in serum of RAS patients with Minor (n=37) or Major (n=35). There were no significant differences between the groups (p<0.4) using unpaired t-test. Mean ± SEM

146

In conclusion, two salivary pro-inflammatory cytokines (IL-6 and IFN-g) were found to be significantly raised in Major RAS patients compared with Minor RAS subjects (p<0.05). This suggests that the severity of the ulcers are directly related to these two specific pro-inflammatory cytokines.

In serum, no cytokines were found to be raised in Major compared with Minor RAS.

4.5 Effects of different modalities of treatments on Cytokines:

This section addresses the question of whether high values of cytokines in (pre-treatment) RAS patients were affected by different modalities of treatments such as topical, systemic or combined of topical and systemic therapies of RAS.

The 72 RAS patients were further divided according to different modalities of treatment into: (25) subjects on Betnesol mouthwash QDS/BD, (22) subjects on Colchicine 0.5mg OD and (25) subjects on Colchicine 0.5mg OD and Betnesol mouthwash QDS. Nine cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g) were analysed in serum and saliva before and after 12 months of treatment.

Four cytokines (IL-8, IL-6, IL-10 and TNF-a) were found to be significantly changed according to treatments (p<0.05), while five cytokines (IL-2, IL4, IL5, IL-17 and IFN-g) were not affected.

Salivary IL-8 levels were reduced after Betnesol treatment:

Betnesol mouthwash QDS/BD : caused a significant reduction in salivary levels of IL-8 from (4518.6 ± 2249.3 pg/ml) to (2912.7 ± 1469.7) (SD) after a treatment period of 12 months (p<0.001). Serum IL-8 concentration reduced from (42.5 ± 28.4) to (37.9 ± 34.1) but was not statistically significant (p<0.3). (Fig.4.32 and 33) 147

Colchicine tablet 500mcg OD : did not cause any significant change either in saliva or in serum. (Fig.4.32 and 33)

Colchicine OD and Betnesol QDS : caused a significant reduction in salivary levels of IL-8 from (3699.2 ± 1752.2) to (2437.4 ± 1298.3) (p<0.001). The mean serum concentration reduced from (105.8 ± 197.2) to (38.6 ± 21.2) which was statistically significant (p<0.05). (Fig.4.32 and 33)

IL-8 (saliva) 6000

5000 * 4000 * 3000 Before Rx 2000 After Rx 1000

0 B C C&B Before Rx 4518.6 3864.7 3699.2 After Rx 2912.7 3634.4 2437.4

Fig.4.32 Comparison of salivary IL-8 in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test *(p<0.001). Mean ± SEM

IL-8 (serum) 250

200

150 * 100 Before Rx After Rx 50

0 B C C&B Before Rx 42.5 148.5 105.8 After Rx 37.9 82.6 38.6

148

Fig.4.33 Comparison of serum IL-8 levels in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test *(p<0.05).Mean ±SEM

Salivary IL-6 levels decreased after treatment with Betnesol:

Betnesol mouthwash QDS/BD: caused a significant reduction in salivary levels of IL-6 from (104.8 ± 213.2 pg/ml) to (38.8 ± 109.2) (SD) (p<0.005). Serum IL-6 concentrations were at the limit of detection in all groups, and no significant changes were found after treatment (decreased from 1.7 ± 0.6 to 1.4 ± 0.5). (Fig.4.34 and 35)

Colchicine tablet 500mcg OD : did not cause any significant change either in saliva or in serum. (Fig.4.34 and 35)

Colchicine OD and Betnesol QDS: caused a significant decrease in salivary IL-6 level from (77.9 ± 129.9) to (26.7 ± 31.6) (p<0.01). Serum IL-6 were at very low levels in all groups, and no significant changes were found after treatment. (Fig.4.34 and 35)

IL-6 (saliva) 160 140 120 100 * 80 * 60 Before Rx 40 After Rx 20 0 B C C&B Before Rx 104.8 39.3 77.9 After Rx 38.8 34.3 26.7

Fig.4.34 Comparison of salivary IL-6 in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test *(p<0.02). Mean ± SEM

149

IL-6 (serum) 2 1.8 1.6 1.4 1.2 1 0.8 Before Rx 0.6 After Rx 0.4 0.2 0 B C C&B Before Rx 1.7 1.7 1.4 After Rx 1.4 1.6 1.4

Fig.4.35 Comparison of serum IL-6 levels in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test. Mean ± SEM

Salivary IL-10 levels were reduced after Betnesol treatment:

Betnesol mouthwash QDS/BD: caused a significant decrease in salivary levels of IL-10 from (8.2 ± 8.5 pg/ml) to (3.8 ± 3.1) (SD) (p<0.002). There was no significant change in mean serum concentrations of IL-10 before and after treatment. (Fig.4.36 and 37)

Colchicine tablet 500mcg OD : no changes were observed either in saliva or in serum. (Fig.4.36 and 37)

Colchicine OD and Betnesol QDS : salivary IL-10 levels were reduced from (8.9 ± 8.2) to (6.4 ± 9.2) but this did not reach statistical significance (p<0.09). Serum concentrations were reduced from (2.7 ± 1.2) to (2.6 ± 1.2) which were not statistically significant (p<0.2). (Fig.4.36 and 37)

150

IL-10 (saliva) 12

8 *

6 Before Rx 4 After Rx 2

0 B C C&B Before Rx 8.2 7.9 8.9 After Rx 3.8 6.8 6.4

Fig.4.36 Comparison of salivary IL-10 in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test *(p<0.002). Mean ± SEM

IL-10 (serum) 7 6 5 4

3 Before Rx 2 After Rx 1 0 B C C&B Before Rx 2.6 4.3 2.7 After Rx 2.4 3.1 2.6

Fig.4.37 Comparison of serum IL-10 levels in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test. Mean ± SEM

151

Salivary TNF-a levels were reduced after treatment with Betnesol:

Betnesol mouthwash QDS/BD: a significant decrease was noticed in salivary TNF-a concentrations from (16.7 ± 27.4 pg/ml) to (8.3 ± 9.7) (SD) (p<0.03). No significant changes were detected in serum levels of TNF-a. (Fig.4.38 and 39)

Colchicine tablet 500mcg OD : no changes were observed either in saliva or in serum. (Fig.4.38 and 39)

Colchicine OD and Betnesol QDS : there were no significant changes either in salivary TNF-a levels, which were reduced from (10.8 ± 15.6) to (7.6 ± 6.3) (p<0.1), or in serum TNF-a concentrations, which were decreased from (3.6 ± 1.1) to (3.5 ±1.4) (p<0.3). (Fig.4.38 and 39)

TNF-a (saliva) 25

15 *

10 Before Rx After Rx 5

0 B C C&B Before Rx 16.7 13.2 10.8 After Rx 8.3 10.3 7.6

Fig.4.38 Comparison of salivary TNF-a in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test *(p<0.03). Mean ± SEM

152

TNF-a (serum) 4.5 4 3.5 3 2.5 2 Before Rx 1.5 1 After Rx 0.5 0 B C C&B Before Rx 4 3.7 3.6 After Rx 3.7 3.7 3.5

Fig.4.39 Comparison of serum TNF-a levels in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test. Mean ± SEM

Salivary IFN-g levels unchanged after treatment:

Betnesol mouthwash QDS/BD: there was no significant reduction in salivary IFN-g levels which were reduced from (12 ± 33.4 pg/ml) to (10.3 ± 32.8) (SD) (p<0.1). In serum, no significant changes were seen before and after treatment in IFN-g concentrations (p<0.3). (Fig.4.40 and 41)

Colchicine tablet 500mcg OD : no changes were observed either in saliva or in serum. (Fig.4.40 and 41)

Colchicine OD and Betnesol QDS : no significant changes were noticed in salivary IFN-g level, which was reduced from (15.1 ± 44.4) to (8.6 ± 17.4) (p<0.2). In serum, no significant changes were seen in IFN-g concentrations (p<0.3). (Fig.4.40 and 41)

153

IFN-g (saliva) 30

15 Before Rx 10 After Rx 5

0 B C C&B Before Rx 12 5.8 15.1 After Rx 10.3 4.9 8.6

Fig.4.40 Comparison of salivary IFN-g in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test. Mean ± SEM

IFN-g (serum) 4 3.5 3 2.5 2 1.5 Before Rx 1 After Rx 0.5 0 B C C&B Before Rx 2.8 3.1 2.6 After Rx 2.7 3.4 2.7

Fig.4.41 Comparison of serum IFN-g levels in (25) RAS patients on Betnesol mouthwash, (22) on Colchicine and (25) on Colchicine and Betnesol, before and after 12 months of treatment using paired t-test. Mean ± SEM

154

In conclusion, treatment with Betnesol mouthwash QDS/BD as a potent anti- inflammatory local steroid resulted in significant reductions in the salivary concentrations of four pro-inflammatory cytokines (IL-8, IL-6, IL10 and TNF-a) (p<0.05), whereas Colchicine tablets 0.5mg OD as an anti-inflammatory systemic therapy had no significant effect on cytokines either in saliva or in serum. Combined treatment of Colchicine tablet and Betnesol mouthwash caused significant reductions in two salivary pro-inflammatory cytokines IL-6 (p<0.05) and IL-8 (p<0.001), though in serum its effect was limited to the reduction of IL-8 only (p<0.05).

In summary, salivary IL-6, IL-8, IL-10, TNF-a and IFN-g concentrations were found to be significantly raised in RAS subjects compared with healthy controls. In addition, salivary IL-6 and IFN-g were significantly raised in the presence of ulcers compared with non-ulcers and in Major RAS compared to Minor. This suggests that salivary IL-6 and IFN-g are directly related to the inflammatory process and to the severity of RAS. Salivary levels of IL-6, IL-8, IL-10 and TNF-a were significantly reduced by the anti-inflammatory effect of local steroid mouthwash (Betnesol), while combined therapy of systemic Colchicine and Betnesol mouthwash had a significant effect on reductions of salivary IL-6 and IL-8 after 12 months of therapy.

In serum, IL-8 was the only cytokine that was found to be significantly raised in RAS subjects compared with controls. Although its concentration was increased in the presence of ulcers compared with non-ulcers and in Major compared to Minor RAS, this was not statistically significant (p<0.09). IL-8 was significantly reduced by the anti-inflammatory effects of the combined treatment of systemic Colchicine and local steroid (Betnesol mouthwash) after 12 months of treatment.

155

CHAPTER FIVE DISCUSSION

Comparison of three different modalities of treatments for the management of Recurrent Aphthous Stomatitis (RAS)

5.1 Complete series:

This randomised, prospective, parallel-group, single-centre clinical trial assessed the efficacy of three different modalities of treatment at 3 month intervals for 12 months on randomly selected RAS patients.

Ulcer Severity Scores (USS) of 86 RAS patients before and after treatment:

The results of our study showed that highly significant reduction in the mean USS score for the 86 RAS patients, from 34.9 (before treatment) to 17.5, with an improvement of 49.9% after 12 months therapy. The mean USS was significantly decreased after 3 months to 24.5, with an improvement of 29.8%. Further reductions were seen after 6 months to 21.1, with an improvement of 39.5%, and after 9 months therapy to 18.8, with an improvement of 46.1%.

Effect of treatments on six individual characteristics of ulcers:

Our study findings showed that the individual ulcer characteristics in the 86 RAS participants, significant improvements were noticed in all six parameters of size, number, duration of the ulcers, pain score, sites affected and the length of the ulcer free periods after 6 months therapy. Further improvements were demonstrated in all ulcer parameters after 12 months treatment, apart from the number of ulcers. Nevertheless, the USS should reveal differential effects of these characteristics in different therapies.

156

USS in Minor and Major RAS before and after treatment:

Our results showed that there was a highly significant reduction in the mean USS score of the 46 Minor RAS subjects from 29.5 (before treatment) to 14.5, with an improvement of 50.8% after 12 months therapy. The mean USS score of the 40 Major RAS participants decreased from 40.8 (before treatment) to 21 after 12 months of treatment with an improvement of 48.5%. However, Minor and Major RAS subjects showed similar responses for the whole groups and most of the improvements were evident by 9 months.

In the current study the USS score for the Minor RAS group ranged from 20 to 37 while for the Major RAS group, it ranged from 33 to 53. This overlap between 33 and 37 was due to the fact that patients attending an oral medicine clinic with Minor RAS are at the more severe end of the severity spectrum and do not represent the expected severity distribution in the population at large. The severity scores for Major RAS would be expected to be greater than for Minor RAS overall, as the ulcers are larger in size, last longer and affect both keratinised and non-keratinised mucosa. This was similar to a study by Tappuni et al (2013) of 223 RAS subjects, where the USS score ranged from 18 to 43 for the Minor RAS group and ranged from 28-60 for the Major group (Tappuni et al 2013).

5.2 Betnesol mouthwash QDS/BD group:

There is a widely held assumption that Betnesol mouthwash is effective. However at present, there are very few published randomised clinical trials over a good timespan to evaluate the efficacy of Betnesol mouthwash for the management of RAS.

157

USS of RAS patients before and after treatment with Betnesol mouthwash:

The results of our study showed a highly significant reduction in the mean USS score for the 29 RAS patients in the Betnesol mouthwash group, from 34.6 (before treatment) to 15.7, representing an improvement of 54.6% after 12 months of therapy. The results of this clinical trial demonstrated that 13.8% of patients (4 out of 29) had no recurrences during the treatment period, 10.3% (3 out of 29) did not show any improvement and 75.9% (22 out of 29) showed more than 20% improvement with no side effects reported for the duration of the follow up period of 12 months. These findings suggest that Betnesol mouthwash QDS during ulcer attacks and BD in between can be an effective treatment for RAS in the majority of the patients.

Our study showed that patients on Benesol mouthwash noted the benefits within the first 3 months of therapy with a significant improvement of 35%. Major reductions in the mean USS were seen at 6 months, with a mean improvement of 47.1% and at 9 months with a mean improvement of 55.8%.

Our study findings were in agreement with a study by Tappuni et al (2013), which included 79 RAS subjects who used Betnesol mouthwash QDS/BD for 3 months. Tappuni et al showed that the USS score was improved by at least 20% in 38% of participants, while 8% had no recurrences during treatment periods and 24% of patients remained unchanged (Tappuni, Shirlaw, Challacombe 2013). However, our study results showed a higher percentage of patients (75.9%) who reported more than 20% improvement. This could be due to the short period of Tappuni et al study of 3 months.

Effect of Betnesol mouthwash on six individual characteristics of ulcers:

RAS subjects on Betnesol mouthwash QDS/BD had significant reductions in all ulcer characteristics (size, number, duration, pain and site) and an increase in the length of ulcer free periods at 6 months of treatment. Although 158 further reductions had been noted in all ulcers’ parameters after 12 months therapy, only the number of ulcers and the length of ulcer free periods continued to improve significantly for 12 months while size, duration, pain and site reductions were not statistically significant.

Our study findings were in agreement with those of Tappuni et al (2008), in a study that included 45 RAS patients who used Betnesol mouthwash for a short period of 3 months, compared with 45 RAS subjects used Betnesol mouthwash and Colchicine tablet 0.5mg/day. The authors reported that five ulcer parameters (size, number, duration, pain and site) were significantly decreased in the Betnesol group while there was no change in the length of ulcer free periods (Tappuni et al 2008). Our study's results showed that the improvement in the length of ulcer free periods was 27.2% after 6 months therapy and 45.2% at 12 months. This was probably due to the established treatment time of the current study.

USS in Minor and Major RAS before and after treatment with Betnesol mouthwash:

The results of our study showed that there was a highly significant reduction in the USS score of the 17 Minor RAS subjects from 29.2 (before treatment) to 13.2 with a mean improvement of 54.7% after 12 months therapy. However, most of the improvements of Betnesol on Minor RAS were evident by 9 months. The mean USS score of the 12 Major RAS participants decreased from 42.1 (before treatment) to 19.1 with an improvement of 54.6% after 12 months therapy and most of the improvement was evident by 6 months. This suggests that Betnesol mouthwash needs to be used for at least 6 months to observe significant improvements.

At present, there are very few published randomised clinical trials to evaluate the efficacy of Betnesol mouthwash for the management of RAS. However, Betnesol mouthwash was used for the management of Lichen planus in a cross-over, randomised study by Hegarty et al (2002). The study compared Betnesol mouthwash QDS with Fluticasone propionate spray for 6 weeks in 159

44 patients, with a washout period of 2 weeks. Their results showed that there was no significant difference between the 2 steroids in pain score reduction (VAS). Betnesol mouthwash reduced the size of the lesions by 41% and the pain score by 49%. Our study showed similar findings, Betnesol caused a reduction in the size of the ulcer by 51.7% and in the pain score by 49.9% after 6 months therapy. Hegarty et al study showed that there were no adverse events reported in the Betnesol mouthwash group. Our study showed similar findings, where no side effects were noticed in patients on Betnesol QDS/BD for the duration of 12 months (Hegarty et al 2002).

In conclusion, the question of the length of time needed to obtain the maximum benefit has not appeared to be addressed in any other clinical trial. Therefore, we suggest that 9 months therapy of Betnesol mouthwash QDS during ulcer attacks and BD in between is recommended to get the maximum benefit of Betnesol for the management of RAS. To our knowledge this is the first randomised controlled clinical trial that has assessed the effectiveness of Betnesol mouthwash over a long timespan where patients were followed for 12 months and reviewed at 3 monthly intervals using USS score plus patient daily diaries and clinical examinations to validate the history given by the patients.

5.3 Colchicine 0.5mg OD group:

Previous studies provided a moderate amount of evidence that supports the effectiveness and the safety of Colchicine for the management of RAS. Nevertheless, the question of the length of time needed to obtain the maximum benefit has not appeared to be addressed in these studies. Colchicine was first used for the treatment of RAS by Ruah et al (1988), for 3 patients with long standing active RAS.

160

USS of RAS patients before and after treatment:

The results of our study showed a highly significant reduction in the mean USS score for the 26 RAS subjects in the Colchicine as monotherapy group, from 32.5 (before treatment) to 17.7 after 12 months therapy. The results of this clinical trial demonstrated that 7.7% (2 out of 26) of patients had no recurrences during the treatment period, 19.2% (5 out of 26) did not show any improvement and 73.1% (19 out of 26) showed more than 20% improvement. However, 11.4% (4 out of 35) developed adverse events of alopecia, gastric pain and vomiting. These findings suggest that systemic Colchicine 0.5mg OD can be an effective treatment for RAS in the majority of the patients. Our study showed that patients on Colchicine noted the benefits within the first 3 months of therapy with a mean improvement of 31.7%.

Effect of Colchicine on six individual characteristics of ulcers:

Our study results showed that patients in this group had a significant reduction in all six ulcer characteristics (size, number, duration of the ulcers, pain and sites affected) and a significant increase in the length of ulcer free periods at 6 months of treatment. Although further reductions had been noted in all ulcers’ parameters after 12 months therapy, only the length of ulcer free periods with a mean improvement of 48.4% continued to improve significantly for 12 months.

USS in Minor and Major RAS before and after treatment with Colchicine 0.5mg OD:

The results of our study showed that there was a highly significant reduction in the mean USS score of the 18 Minor RAS subjects from 28.8 (before treatment) to 14.4 after 12 months therapy. However, most of the improvements of Colchicine on Minor RAS were evident by 3 months. The mean USS score of the 8 Major RAS participants decreased from 40.6 (before treatment) to 25, most of the improvement on Major RAS was evident

161 by 3 months. This suggests that Colchicine needs to be used for at least 3 months to observe significant improvements.

Similar findings were demonstrated in an open clinical trial by Hassan et al (2010), where 30 RAS patients received Colchicine 1.2mg a day for 2 months. Hassan et al showed a significant reduction in pain score by 82.5% (83.2% in Minor, 78.5% in Major and 90% in Herpetiform RAS). In addition, significant improvements in aphthous counts of 87.3% were reported (88.3% in Minor, 85.7% in Major and 83.3% in Herpetiform RAS). Adverse events were reported in 46.7% of patients: 26.7% (abdominal pain), 13.3% (diarrhoea) and 6.7% (vomiting). Our study findings were in agreement with Hassan et al, but our study showed a lower percentage of improvement, where an improvement of 50.1% was seen in the Minor RAS group and 38.5% in the Major RAS group after 12 months treatment with Colchicine 0.5mg OD. Our study showed a significant reduction in pain by 30.6%, an improvement in the number of ulcers by 48.6% after 6 months of therapy and fewer side effects reported (11.4%). This could be due to the low dose of Colchicine. However, our study showed that an improvement of 28.5% was seen in 10 Major RAS patients on Colchicine 1mg/day, who were recalcitrant to treatment with Colchicine 0.5mg/day.

Our study findings with a similar dose of Colchicine were in agreement with those of Pakfetrat et al (2010) in a double blind, randomised, clinical trial which included 34 RAS patients who received a daily dose of either 0.5 mg of Colchicine (n=17) or 5mg/day of Prednisolone (n=17) over a short period of 3 months. Pakfetrat et al showed that patients in the Colchicine group reported a significant pain reduction and burning sensation of more than 85% during a 3 month period of therapy. Adverse events of gastrointestinal disturbances, headache and vertigo were reported in 52.9% of the Colchicine group, which was higher than for the Prednisolone group (11.8%): as a result the authors suggested that Prednisolone was a safer choice in the management of RAS. However, this study did not consider the long-term side effects of systemic steroids. Moreover, all patients were not on any medications for 2 weeks prior to their inclusion in the trial, it is questionable whether this was a long enough 162 wash out period (Pakfetrat et al 2010). Our study findings were in agreement with the Pakfetrat et al study in showing reductions in pain score by 30.6% after 6 months and by 43.2% after 12 months therapy with the same low dose of Colchicine (0.5mg/day), but in our study there were fewer adverse events reported (11.4%) of gastrointestinal disturbances and alopecia. This could be due to the short wash out period (2 weeks) in Pakfetrat et al study while in our study, it was 3 months.

Further agreement with Lynde et al (2009), but with a higher dose of Colchicine (1.2-1.8mg/day), in a retrospective study which included 50 RAS patients. Lynde et al reported that 3% of patients showed no recurrences during the treatment period, 26% had not responded to the therapy and 60% showed an improvement of 75% in the frequency, intensity and severity of the ulcers. However, the limitation of this study was the ambiguity in assessing the response to treatment. Adverse events of diarrhoea were seen in 31% of the patients. Our study showed that 7.7% of RAS patients had no recurrences during treatment periods, 19.2% were a non-responder to therapy and 73.1% reported more than 20% improvement after 12 months treatment with Colchicine 0.5mg/day. These findings were in agreement with Lynde et al study. However, adverse events were reported by only 11.4% of patients in our study, while in Lynde et al 31% reported side effects; this could be related to the high dose of Colchicine used in their retrospective study (1.2- 1.8mg/day).

Our study findings were in agreement with a study by Mimura et al (2009), who demonstrated the effectiveness of a high dose of Colchicine (1.5mg a day) in a small number of 10 RAS patients who were followed for 2-6 months. 90% of patients showed a significant reduction of pain, 40% had no recurrences during the treatment period and 30% experienced gastrointestinal disturbances (diarrhoea). However, the absence of randomisation and blinding, as well as the relatively small sample, could limit the validity of this study and necessitates caution when interpreting their results. Our study showed similar findings for the reductions of pain by 30.6% after 6 months of treatment and by 43.2% after 12 months, 7.7% had no recurrences of the 163 ulcers during the treatment period, but in our study there were less side effects reported (11.4%), this could be due to the high dose of Colchicine used in Mimura et al study.

Our study findings were in agreement with those of Fontes et al (2002), in a study that included 54 patients with severe RAS. Fontes et al suggested that Colchicine in a dose of 1-1.5mg a day for 3 months prevented recurrences of ulcers in 22% of patients, while 41% of patients reported a significant improvement of 50% in frequency, duration of ulcers and pain. Although patients were followed for a mean of 4.7 years (from 6 months to 13 years), all of the follow ups were done by phone only (Fontes et al 2002). Our study showed similar findings with reductions in the duration of ulcers by 53.6% and pain score by 43.2% with an increase in the length of the ulcer free periods by 48.4% after 12 months therapy with a low dose of Colchicine (0.5mg/day).

In summary, there is a moderate amount of evidence that supports the effectiveness and the safety of Colchicine for the management of RAS. Nevertheless, the question of the length of time needed to obtain the maximum benefit has not appeared to be addressed in the previous studies. Our study suggest that 3 months therapy of Colchicine 0.5mg OD is recommended to get the maximum benefit of Colchicine with a minimum adverse events for the management of RAS. To our knowledge this is the first randomised controlled clinical trial that assessed the effectiveness of Colchicine as monotherapy over a long timeframe where patients were followed for 12 months and reviewed on 3 monthly intervals using a standardised method (USS) plus patient daily diaries and clinical examinations to validate the history given by the patients.

5.4 Colchicine 0.5mg OD and Betnesol mouthwash QDS group:

There is preliminary evidence that combined treatment of Colchicine plus Betnesol mouthwash has a role to play in the management of RAS.

164

USS before and after treatment with Colchicine and Betnesol:

The results of our study showed a highly significant reduction in the mean USS score for the 31 RAS patients in the Colchicine plus Betnesol group, from 36.9 (before treatment) to 19.2 after 12 months therapy. The results of this clinical trial demonstrated that 3.2% (1 out of 31) had no recurrences during treatment period, 12.9% (4 out of 31) did not show any improvement and 83.9% (26 out of 31) showed more than 20% improvement with Colchicine 0.5 mg OD plus Betnesol mouthwash QDS during ulcer attacks for the duration of 12 months. However, 5.6% (2 out of 36) developed adverse events of peripheral neuropathy and gastrointestinal disturbances (nausea) after 3 months. These findings suggest that systemic Colchicine 0.5mg OD plus Betnesol mouthwash QDS can be an effective treatment for RAS in the majority of the patients.

Our study showed that patients on Colchicine plus Benesol noted the benefits within the first 3 months of therapy with a significant improvement of 25.2%. Major reductions in the mean USS were seen at 6 months, with an improvement of 37.4% and at 9 months, with an improvement of 45.3%.

Effect of Colchicine and Betnesol treatment on six individual characteristics of ulcers:

RAS subjects on combined therapy of Colchicine 0.5mg OD and Betnesol mouthwash QDS had significant reductions in all ulcer parameters (size, number, duration, pain and site) and an increase in the length of ulcer free periods at 6 months of treatment. Although further improvements had been noted in all ulcers' characteristics at 12 months therapy, only the size of ulcer, pain score and the length of ulcer free periods continued to improve significantly for 12 months. Our study findings were comparable to a study by Tappuni et al (2008). Tappuni et al reported that Colchicine plus Betnesol for 3 months in 45 RAS patients caused significant reductions in the size of ulcers, number, duration, pain score, and sites affected as well as an increase in the length of ulcer free 165 periods. Our study had similar findings for the first 6 months, but since we followed the patients for 12 months, we demonstrated that only size, pain and ulcer free period, were significantly in constant improvement; number, site and duration of the ulcers were not statistically significant (p<0.06).

USS in Minor and Major RAS after treatment with Colchicine and Betnesol:

The results of our study showed that there was a highly significant reduction in the mean USS score of the 11 Minor RAS subjects from 31 (before treatment) to 16.6 with after 12 months therapy. However, most of the improvements of Colchicine plus Betnesol on Minor RAS were evident by 6 months. The mean USS score of the 20 Major RAS patients decreased from 40.2 (before treatment) to 20.6, with a mean improvement of 48.7% after 12 months therapy. Most of the improvements on Major RAS were evident by 9 months. This suggest that this combined therapy of Colchicine and Betnesol mouthwash needs to be used for at least 6 months to observe significant improvements.

At present, there are very few published randomised clinical trials to evaluate the efficacy of Colchicine plus Betnesol mouthwash for the management of RAS. Nevertheless, combined treatment of Colchicine plus systemic steroids were demonstrated in a study by Viguier et al (2000), who studied the effect of 2mg a day of Colchicine plus Prednisolone 1mg/kg in 5 cases of Herpetiform RAS. Although the study reported that Prednisolone resulted in faster healing of ulcers while Colchicine managed pain and increased the length of ulcer free periods, their sample size was very small and consisted of men only. Similar findings were seen in our study, which showed that treatment with a low dose of Colchicine plus local steroids for 12 months caused reductions in pain score by 47.5%, duration of the ulcers by 53.7% and increased the length of ulcer free period by 36.1%.

Comparable findings were seen in a retrospective study by Letsinger et al (2005), which included 42 patients with complex aphthosis. Letsinger et al 166 reported that Colchicine 1.8mg/day combined with Dapsone 100mg/day had a superior effect (57% of patients achieved 50% improvement) than that of Colchicine as monotherapy (28% of patients achieved 50% improvement) in managing complex aphthosis. Despite the retrospective design of this study, the vagueness in response to treatment was one of their major limitations. Our study showed that 83.9% achieved more than 20% improvement using combined treatment of Colchicine plus Betnesol.

Our study showed similar findings to those of Genvo et al (1984), in a study which used a very high dose of Colchicine 3mg/day plus Thalidomide 300mg/day in 8 patients with RAS. Although major improvement has been noted, with a rapid healing of the lesions and reductions of pain and burning sensation, due to the small sample size no conclusions were drawn with regard the efficacy of the treatment. Despite the high dose used in this open clinical trial, Colchicine did not prevent recurrences as new lesions were reported by the patients a few weeks after discontinuation of the treatment (Genvo et al 1984). Our study findings showed that treatment with a low dose of Colchicine plus Betnesol for 12 months caused reductions in pain score by 47.5% and in the duration of ulcers by 53.7%, as well as increasing the length of ulcer free periods by 36.1%.

In summary, our study suggest that 9 months therapy of Colchicine 0.5mg OD and Betnesol mouthwash QDS during ulcer attacks only is recommended to obtain the maximum effect of this combined therapy for the management of RAS. To our knowledge this is the first randomised controlled clinical trial that assessed the effectiveness of Colchicine plus Betamethasone mouthwash over a long timeframe, where patients were followed for 12 months and reviewed on 3 monthly intervals using a standardised assessment (USS) plus patient daily diaries and clinical examinations to validate the history given by the patients.

167

The results of this prospective long term clinical trial showed that 6 out of 71 (8.5%) RAS patients, who were treated with Colchicine 0.5mg a day with or without Betnesol mouthwash developed adverse events. Four patients experienced gastrointestinal disturbances (diarrhoea and vomiting), one patient developed alopecia and one patient reported peripheral neuropathy which was reversible. However, gastrointestinal disturbances were reported in most of the published literatures while alopecia was noted in a case report by Harms in 1980, in which a 43 year old Female experienced alopecia after 2 months treatment with a low dose of Colchicine (Harms 1980). In addition, peripheral neuropathy was reported by Choi et al (1999) in a case report of a 84 year old Female who developed a reversible peripheral neuropathy and myopathy with Colchicine 0.5mg BD, as was confirmed by Electromyography test (EMG) (Choi et al 1999).

There is moderate evidence that supports the superiority of Colchicine in terms of safety and efficacy. A study by Zemer et al in 1991 which included 350 children aged less than 16 years old with FMF (Familial Mediterranean Fever), who were treated over a long term (6-13 years) with Colchicine 1- 2mg/day, highlighted the safety of this drug. The side effects experienced with Colchicine were mostly diarrhoea and nausea but did not require permanent discontinuation of the treatment for any of the children (Zemer et al 1991). This finding was supported in a more recent study by Tasher et al in 2008 which included 9 children with PFAPA (Periodic Fever, Aphthous Stomatitis, Pharyngitis, and Adenitis), aged 3.5-11.0 years, who were treated with Colchicine at a daily dosage of 0.5-1.0 mg and were followed for 2 years. Only one patient reported abdominal pain (Tasher et al 2008). These findings were supportive to our study results where 8.5% of patients (6 out of 71) on a low dose of Colchicine developed adverse events of gastrointestinal disturbances (vomiting and diarrhoea), alopecia and peripheral neuropathy which necessitated termination of their medication.

Of the 86 RAS patients in this study, 45.3% were Male and 54.6% were Female (a ratio of 1 :1.2). The patients had a mean age of 39.8 years old and 83% were white. This gender and racial distribution is consistent with 168 epidemiological studies of RAS (Bhatnagar et al 2013, Fahmy 1976), although the mean age in our study was slightly older.

In this study, 53.5% (46 out of 86) of patients had Minor RAS and 46.5% (40 out of 86) had Major. This high percentage of Major RAS than Minor subjects in this study were in agreement with Tappuni et al (2013), in a study which included 79 RAS subjects with 41 (51.9%) had Minor and 38 (48.1%) had Major RAS. However, it differed from Bagan et al (1991), in a study of the clinical characteristics of 93 RAS patients, of whom 70.9% had Minor, 21.5% had Major and 7.5% had Herpetiform RAS. This is certainly due to our subjects being specialist clinic attendees with a more severe form of RAS who needed to be referred to a tertiary clinic.

One of the limitations in the present study is the lack of a placebo controlled group, which was due to the study’s long timespan of 12 months and the fact that patients with severe types of RAS in particular could not be left without treatment for the long period of the clinical trial. Another limitation of this study is the lack of blinding in clinicians and participants, which was due to the differences in the mode of the applications since one is a tablet and the other is a mouthwash. Subjects with Herpetiform RAS were not included in the study because we did not have sufficient enrolment from this group to allow statistical comparisons. None of the patients were on any form of treatment for their RAS for at least 3 months prior to their inclusion in the trial. The number of participants was lower than anticipated due to the strict inclusion and exclusion criteria. All patients who were included in the study were strictly RAS and did not have haematological deficiencies, gastrointestinal disorders or Behcet disease. All patients had normal blood test results prior to the inclusion in the trial. The age of the participants was limited to between 18 to 65 years due to the possibility of adrenal suppression as a side effect of steroids in children, and due to the Colchicine adverse events of gastric bleeding, renal and hepatic impairment in those older than 65 years old. Of course, pregnant women and breast feeders were excluded because of the teratogenic effect of Colchicine and the adrenal suppression of steroids. Any participants who were not able to understand verbal explanations or written 169 information given in English were excluded to prevent the study outcome being affected by the misinterpretation of information and instructions given. In addition, all patients in this study were not involved with other studies in order to not compromise their safety as well as to prevent drugs interactions. All patients with a past history of allergy, hypersensitivity, cardiac, renal, hepatic, gastrointestinal diseases or blood dyscrasias were excluded to insure patient safety and to prevent drug interactions and Colchicine cytotoxicity, which might cause gastric bleeding, renal and hepatic impairment. All subjects had not been on systemic steroids, Colchicine or Betnesol, for at least 3 months prior to the inclusion. This was to prevent interaction effects from other types of steroids that might have affected the evaluation of the study outcome.

In conclusion, the results of the present study suggest that 9 months treatment with Betnesol mouthwash QDS during ulcer attacks and BD in between could be used as a first line standard and safe therapy for the management of RAS. It was effective in reductions of size, number, duration, pain and site and has the potential to prevent recurrences during the treatment period in the majority of RAS patients with no side effects reported. The effects of Betnesol were equivalent for Minor and Major RAS.

Our study findings showed that the main benefit of a low dose (0.5mg/day) of Colchicine as monotherapy was over the first 3 months and little further improvement was seen thereafter. The effect of Colchicine on Minor RAS (with an improvement of 50%) was greater than Major RAS (with an improvement of 38.5%). Therefore, we suggest that treatment with 1mg/day of Colchicine was more effective for recalcitrant RAS patients.

The results of our study suggest that 9 months treatment with Colchicine 0.5mg OD and Betnesol mouthwash QDS during ulcer attacks for Major RAS. It was effective in reductions of size, number, duration, pain and site and has the potential to prevent recurrences during the treatment period in the majority of RAS patients with few adverse events reported.

170

CHAPTER SIX DISCUSSION

Cytokines profile of Recurrent Aphthous Stomatitis (RAS) and the effects of treatments on serum and salivary cytokines

6.1 Cytokines in Healthy controls and RAS patients:

The purposes of this study were: a) to determine in cross sectional studies if cytokines in serum or saliva are altered in RAS and b) in longitudinal studies to determine whether local (Betnesol) or systemic (Colchicine) therapies results in changes in serum or salivary cytokines and whether these changes might be related to successful treatment of RAS.

One of the possible factors involved in the pathogenesis of RAS is a cell- mediated immune response in which several cytokines seem to play a major role. Our study findings showed that salivary concentrations of IL-6, IL-8, IL- 10, TNF-a and IFN-g were significantly greater in 72 RAS subjects than in 34 healthy controls, while salivary levels of IL-2, IL-4, IL-5 and IL-17 were not significantly different between the groups. In serum, IL-8 was the only cytokine which was detected significantly raised in RAS subjects compared with healthy controls, though in a level that was much lower than salivary concentrations. These findings suggest a role for the pro-inflammatory cytokines in the pathogenesis of RAS. However, previous studies on the interrelationship of these cytokines in serum and salivary samples of RAS subjects have shown inconsistent results (Sun et al 2000).

Our study findings of raised salivary concentrations of TNF-a in RAS patients compared with controls were in agreement with Valle et al (2011), who demonstrated that salivary TNF-a levels (using Chemiluminescent immunoassay) were significantly higher in 20 RAS patients compared with 10 healthy controls, who were randomly selected (p<0.01). However, this study

171 did not classify their RAS subsection into Minor, Major or Herpetiform to determine whether TNF-a levels were higher in one type more than others (Valle et al 2011).

Our study results showed similar findings to Boras et al (2006) in a study which included 26 RAS subjects and 26 healthy controls (using ELISA). Boras et al showed that salivary levels of TNF-a were significantly raised in RAS patients with the highest levels seen during remission periods (p<0.001). The authors suggested that these findings might be an indication of the possible role of TNF-a in the healing of RAS, but interpretation of these results might be affected with the small sample size of 13 RAS patients. In addition, Boras et al (2006) found that salivary IL-6 did not show any significant differences between RAS patients and controls. This was in contrast with our findings of high concentrations of salivary IL-6 in RAS subjects compared with healthy controls which could be due to that TNF-a exerts its activities with other cytokines especially with IL-6.

The results in the current study showed that there were no significant differences in serum levels of TNF-a in RAS subjects compared with healthy controls. This was in agreement with Thornhill et al (2007), who reported that there were no significant differences in TNF-a plasma concentrations of 19 RAS subjects, since TNF-a was within normal levels in all stages of the disease. The secretion of TNF-a was thought to play a role in the pathogenesis of RAS and to be inhibited by Pentoxyphiline. Thornhill et al showed that plasma levels of TNF-a (detected by ELISA) were similar in pre- treatment, mid-treatment and placebo groups and unchanged even when treatment was successful.

In a histological study, Natah et al (2000) reported a high expression of TNF-a (determined by the intensity of staining) in lesional biopsies of 12 RAS patients compared with 10 controls. However, all subjects were Minor RAS only. Sun et al (2006) in a study that included 146 RAS patients and 54 controls (using Chemiluminescent assay) showed that serum levels of TNF-a were found to be higher than normal in 29% of RAS patients in apparent 172 contrast to our study and to those of Thornhill which could be due to using different techniques may give different results.

In serum, the results of our study demonstrated raised levels of IL-8 in RAS subjects. This was in agreement with a study by Sun et al (2004), which included 146 RAS patients and 54 healthy controls (using Chemiluminescent assay). The authors showed that the serum level of IL-8 was higher than normal in 60% of RAS subjects. However, in contrast to Sun et al, who found that the serum levels of IL-6 were higher than normal in 25% of RAS patients (Sun et al 2004). Our results showed that mean serum concentrations of IL-6 were not significantly increased in RAS subjects.

In contrast to the results of Sun et al, Albanidou-Farmaki et al (2007) showed in a study which included 32 RAS patients and 40 controls (using ELISA) that there were no differences in serum levels of IL-6, TNF-a and IL-5 between RAS patients and controls (p>0.05). These results supported our findings of no differences in serum levels of IL-6, IL-5 and TNF-a between RAS subjects and controls. In addition, Albanidou-Farmaki et al found that there was a significant increase in serum levels of IFN-g, IL-2 and IL-10 (p<0.05) and a significant decrease in IL-4 (p<0.001). This was in contrast with our results of no significant differences in serum concentrations of IFN-g, IL-2, IL-4 and IL- 10 in RAS patients and controls as IL-2 and IL-5 were detected at very low levels.

The results of our study showed that there were no significant differences in serum concentrations of IL-2, IL-4, IL-5, IL-10, TNF-a and IFN-g in RAS subjects and controls. Buno et al (1998) in histological study compared lesions of 21 RAS patients with 7 healthy controls using PCR assay. Buno et al showed that a significant increase in the levels of IL-2 and TNF-a mRNA were detected in lesional mucosa of RAS subjects compared with healthy controls. In addition, decreased levels of IL-10 mRNA were noted in RAS subjects compared with controls, while concentrations of IL-4 and IL-5 mRNA were significantly increased. However, there were only 7 subjects in the

173 control group and larger numbers of participants might have revealed more significant differences (Buno et al 1998).

In serum, the results in the current study showed that IL-2, IL-4, IL-5, IL-10, IL-6, TNF-a and IFN-g were found in few samples at low levels using the Luminex Multi-Analyte Profiling System (MAP). This was in agreement with a study by Wallace et al (2005) who used the Luminex Multiplex Bead Analysis System to measure nine cytokines in serum samples of 104 patients with Behcet disease, 20 RAS patients and 15 healthy controls. This study revealed that IL-4, IL-8, IL-10 and TNF-a were not detected in serum samples of RAS patients, while IL-2, IL-6 and INF-g were found in few samples at low levels and no significant differences were found(Wallace et al 2005). Their results supported our study findings apart from IL-8, which was significantly different from controls in our RAS subjects.

Comparable results to our findings and to those of Wallace et al (2005), using the same method of analysing cytokines (Luminex Multiplex Bead Analysis System), were seen in a study by Curnow et al (2008). The study of 52 patients with Behcet disease, 20 RAS patients and 15 controls showed that serum concentrations of IL-6, TNF-a and CXCL-8 in RAS patients were detected at very low levels, while IL-4 and IL-10 were not detected in any samples and IFN-g and IL-2 were found in a few samples only. Although we used the same analysing method (Luminex Multiplex Bead Analysis System), our results contrasted with those of Wallace et al and Curnow et al in terms of IL-8 findings. IL-8 was found in at very low levels in the Curnow et al (2008) study and it was not detected in the Wallace et al (2005) study, but IL-8 concentrations in serum were found to be significantly higher than controls in our study.

Unfortunately, analysing cytokines using different techniques may give different results and low numbers of RAS participants in these studies might affect the interpretation of their results; a larger sample size might give different findings. Pekiner et al (2012) used Flowcytometry to analyse serum levels of IL-2 and IL-6 in 30 patients with Behcet disease, 30 RAS subjects 174 and 15 controls. The study showed that there were no significant differences in serum levels of IL-2 and IL-6 between the groups. This was similar to our study findings but in contrast to those of Sun et al (2000) (using ELISA), who demonstrated that plasma concentrations of IL-2 significantly increased in 34 RAS patients compared with 32 healthy controls (p<0.05) (Sun et al 2000).

Our study findings showed that IL-8 concentrations in serum were significantly greater in RAS subjects than controls, while levels of IL-2, IL-5, IL-6, IL-10 and IL-4 were very low and were not significantly different between the groups. Similar findings for IL-8 and IL-4 were seen in a study by Lewkowicz et al (2005) of 10 RAS patients compared with 12 healthy controls. The authors reported an increase in the production of Type 1 cytokines IL-2, TNF- a and IFN-g along with IL-5, IL-6 and IL-8 by the peripheral blood mononuclear cells of RAS patients, while anti-inflammatory cytokine IL-10 decreased and there were no significant differences in IL-4 levels between the groups. However, the sample size was very small compared with ours of 72 RAS subjects and 34 controls as well as Lewkowicz et al findings were in vitro which might be not applicable in vivo .

Our study findings have not shown any significant differences in serum concentrations of IFN-g, TNF-a, IL-10 and IL-4 either in RAS patients or in healthy controls. This was in agreement with Dalghous et al (2006), regarding IL-4 and IL-10 but in contrast to their findings regarding IFN-g and TNF-a. The Dalghous et al study demonstrated an increased expression of pro- inflammatory and anti-inflammatory cytokines in 25 patients with Behcet disease, while only pro-inflammatory cytokines were noted in 19 RAS patients. Raised levels of IFN-g and TNF-a were seen in lesional biopsies of RAS subjects, while IL-4 and IL-10 were detected in Behcet disease patients but not in RAS subjects (Dalghous et al 2006).

The results of our study did not reveal significant differences in concentrations of IL-17 either in serum or salivary samples of RAS subjects when compared with controls. These findings were in general agreement with Lewkowicz et al (2011), who conducted a gene expression profile of Th1, Th2, Th3 and Th17 175 using PCR array to compare the relative expression of genes involved in the ulcer lesions of 15 RAS patients and 12 healthy controls. This study showed that the majority of genes up-regulated in RAS lesions were Th1, and that Th2, Th3 and Th17 genes were not found in RAS. Their results showed that the ulcer formation were mainly dependent on the activation of the Th1 immune response, as IL-6 and IL-15 were found to be up-regulated in aphthous lesions. IL-4, IL-5, and IL-10 were not different from controls. IL-17 and IL-23 showed unchanged expression in RAS (Lewkowicz et al 2011). Therefore, it can be concluded that there is little evidence for the involvement of Th17 in RAS. our study findings and those of Lewkowicz et al appear to be in contrast to Al- Samadi et al (2013), who showed in lesional biopsies from 5 RAS patients and 5 controls, that IL-17C staining was significantly stronger in RAS lesions than controls (p<0.006). The Al-Samadi et al study suggested that human oral epithelial cells could produce and respond to IL-17C, a pro-inflammatory cytokine that could stimulate IL-8 and TNF-a productions. However, their control samples were obtained from gingival mucosa during wisdom teeth extractions. Therefore, local inflammation could not be completely excluded which might affect the interpretation of their results (Al-Samadi et al 2013). Nevertheless, IL-17 was mostly related to Behcet disease with active uveitis (Wei Chi et al 2008). Thus although, RAS might be considered as part of a broad spectrum of clinical disease ranging from Minor RAS to Behcet disease, there is only weak evidence so far of any TH17 involvement in RAS.

6.2 Cytokines in relation to the presence of ulcers:

Previous studies showed that raised concentrations of cytokines could be related to the presence of ulcer lesions. Therefore to examine this possibility, our study samples of 41 subjects who had ulcers on the time of collection of the samples were compared with 31 subjects who had no ulcers and 34 healthy controls.

176

The results of our study showed that salivary IL-6 and IFN-g were significantly raised in the presence of ulcers compared with the non-ulcer group of RAS patients. In addition, raised levels of salivary IL-8, IL-10 and TNF-a were seen in the presence of ulcers compared with healthy controls. No significant differences were demonstrated in salivary IL-2, IL-4, IL-5, and IL-17 between the groups. In serum, none of the nine cytokines were found to be significantly different between the groups. The results of this study suggest that two salivary pro-inflammatory cytokines (IL-6 and IFN-g) are directly related to the inflammatory process of RAS lesions, while IL-8, IL-10 and TNF-a were raised in RAS subjects irrespective of whether ulcers were present or not.

These findings are in agreement with Valle et al (2011), who showed that salivary concentrations of TNF-a were significantly higher in the ulcerative stage of 20 RAS patients compared with 10 healthy controls. Our study showed comparable results in some cytokines to a study by Boras et al (2006), who revealed that salivary levels of TNF-a were significantly raised in acute and remission stages of RAS subjects compared with healthy controls. The highest concentrations of salivary TNF-a were seen during the remission period (p<0.001). In addition, Boras et al reported that salivary levels of IL-6 did not show any significant differences between the groups, which was in contrast with our findings of salivary IL-6.

Contradictory results to our findings were demonstrated in a study by Sun et al (2006) that included 146 RAS patients and 54 healthy controls. Raised serum concentrations of TNF-a were found in both exacerbation and post exacerbation stages of RAS lesions. However, TNF-a was found to be greater than normal in 29% of RAS patients only. Another study by the same authors (using ELISA) demonstrated that plasma concentrations of IL-2 significantly increased in the early active stage, late active and exacerbation stage of 34 RAS patients compared with 32 healthy controls (Sun et al 2000). Our study results were in contrast to both Sun et al studies, since no significant differences in serum concentrations of IL-2 and TNF-a were demonstrated in RAS lesion.

177

Similar findings in some cytokines but not for the others were seen in a study by Albanidou-Farmaki et al (2007), which included 32 RAS patients in the active phase of the disease compared with 40 healthy controls (using ELISA). Albanidou-Farmaki et al showed that there were no significant differences in serum concentrations of IL-6, TNF-a and IL-5 in RAS subjects compared with controls (p>0.05), while there was a significant increase in IFN-g, IL-2 and IL- 10 (p<0.05) and a significant decrease in IL-4 (p<0.001). The Albanidou- Farmaki et al study was in agreement with our study's findings of no differences in serum levels of IL-6, IL-5 and TNF-a between ulcer, non-ulcer RAS subjects and controls. However, it was in contrast with our findings of no significant differences in serum concentrations of IFN-g, IL-2, IL-4 and IL-10 in both groups of RAS patients and controls (Albanidou-Farmaki et al 2007).

Our study findings of no significant differences in serum concentrations of IL- 4, IL-5, IL-6 and IL-10 in the presence of ulcer compared with healthy controls were in partial agreement with a study by Lewkowicz et al (2011). Their results showed that levels of IL-4, IL-5 and IL-10 were not significantly different from controls, while level of IL-6 was significantly higher in ulcer lesions of 15 RAS patients compared with 12 healthy controls. Similar results to Lewkowicz et al findings, but in contrast to our results were demonstrated in a study by Sun et al (2003), which included 197 RAS subjects and 77 healthy controls. Their study showed high concentrations of serum IL-6 in exacerbation and post-exacerbation stages of RAS lesions. These results were in contrast to our findings where no differences in serum concentrations of IL-6 were seen in the presence of ulcers compared with the non-ulcer group.

Our study findings were in contrast to a study by Lewkowicz et al (2005) that included 10 RAS subjects and 12 healthy controls. Lewkowicz et al study demonstrated an increased cytokines concentrations of IL-2, TNF-a, IFN-g, IL-5, IL-6 and IL-8, while there were no significant differences in IL-4 levels between the groups. Our study findings showed that no significant differences were seen in serum levels of IL-2, TNF-a, IFN-g, IL-5, IL-6 and IL-8 between the groups. Nevertheless, our study showed similar results with Lewkowicz et 178 al study regards IL-4. However, Lewkowicz et al study sample size was very small as only 10 patients were available in the acute phase of RAS (first 2 days) and only 8 subjects were available in the remission phase of RAS (Lewkowicz et al 2005).

In contrast to our findings where no significant differences were found in serum levels of IL-2, IL-10, IFN-g and TNF-a in the presence of RAS lesions compared with non-ulcer group were demonstrated in a histological study by Buno et al (1998). This study found that increased levels of IL-2, IFN-g and TNF-a mRNA were detected in lesional mucosa of RAS subjects compared with non-ulcer RAS patients. However, only IFN-g level was significantly raised in terms of statistics.

Contradictory results for some cytokines, but in agreement with others were found in a study by Dalghous et al (2006). Raised levels of INF-g and TNF-a were seen in lesional biopsies of 19 RAS subjects, while IL-4 and IL-10 were not significantly different from controls. Their findings were in contrast to our results for IFN-g and TNF-a, but both studies were in agreement for IL-4 and IL-10 (Dalghous et al 2006). Our study demonstrated contradictory results to those of Al-Samadi et al (2013). Their findings showed that the staining of IL-8 and TNF-a were stronger in lesional biopsies of 5 subjects with RAS lesions than 5 controls (p<0.04), while in our study there were no significant differences in serum levels of IL-8 and TNF-a, found between the groups (Al- Samadi et al 2013).

6.3 Comparison of cytokines in Minor and Major RAS:

The results of our study showed that salivary IL-6 and IFN-g were significantly raised in Major RAS compared with Minor RAS subjects. In addition, raised levels of salivary IL-8, IL-10 and TNF-a were seen in the Major and Minor groups compared with healthy controls. No significant differences were demonstrated in salivary IL-2, IL-4, IL-5, and IL-17 between the groups. In serum, none of the nine cytokines were found to be significantly different 179 between the groups. The results of this study suggest that two salivary pro- inflammatory cytokines (IL-6 and IFN-g) are directly related to the severity of the disease, while IL-8, IL-10 and TNF-a were raised in RAS subjects irrespective to the severity of RAS.

Similar findings were seen in a study by Boras et al (2006). Boras et al study demonstrated high concentrations of salivary TNF-a in 26 Minor RAS patients compared with 26 healthy controls. Nevertheless, our study findings of high concentrations of salivary IL-6 in Minor and Major RAS compared with controls were in contrast to Boras et al study, who found that IL-6 did not show any significant differences between Minor RAS and healthy controls. However, the lack of comparison to Major RAS might affect the interpretation of their results (Boras et al 2006).

Our study findings showed that there were no significant changes in the serum levels of TNF-a between Minor and Major RAS groups. This was in contrast to a study by Sun et al (2006), who demonstrated raised concentrations of TNF-a in the serum samples of 146 RAS patients compared with 54 healthy controls. Raised serum concentrations of TNF-a were found in 20% of Minor, 39% of Major and 25% of Herpetiform RAS. However, TNF-a was found to be higher than normal in 29% of 146 RAS patients (Sun et al 2006). In another study, the same authors with the same number of RAS patients (146) and controls (54), reported that serum level of IL-8 was higher than normal in 59% of Minor, 59% of Major and 63% of Herpetiform RAS subjects (Sun et al 2004). In addition, the study found that serum level of IL-6 was higher than normal in 25% of RAS patients, including 19% of Minor, 33% of Major and 25% of Herpetiform RAS patients. However, all their comparisons of different types of RAS were with healthy controls and not with each other as in our study, where we compared Minor with Major RAS. Our study findings of no significant differences in serum levels of TNF-a, IL-6 and IL-8 between the groups were in contrast to the Sun et al results.

In contrast to the results of Sun et al studies, Albanidou-Farmaki et al (2007) showed that there were no differences in serum concentrations of IL-6, TNF-a 180 and IL-5 in 32 RAS subjects compared with controls, while there was a significant increase in IFN-g, IL-2 and IL-10 and a significant decrease in IL-4 (p<0.001). Our study findings of no differences in serum levels of IL-6, IL-5 and TNF-a between Minor, Major RAS subjects and controls were in agreement with Albanidou-Farmaki et al. However, it was in contrast in IFN-g, IL-2, IL-4 and IL-10.

Similar findings to our study results, but in contrast to Albanidou-Farmaki et al, in some cytokines, were seen in a study by Dalghous et al (2006). Raised levels of TNF-a and INF-g were seen in lesional biopsies of 19 Minor RAS subjects, while there were no significant differences detected between RAS subject and controls in IL-10 and IL-4. The results of our study showed that there were no significant differences between the groups in serum levels of TNF-a, IFN-g, IL-10 and IL-4. Our findings contradicted their results for TNF-a and IFN-g, but were in agreement for IL-4 and IL-10 (Dalghous et al 2006).

Contradictory findings were seen in a study by Natah et al in 2000, which reported a high expression of TNF-a in lesional biopsies of 12 Minor RAS subjects compared with 10 healthy controls (Natah et al 2000). In addition, a study by Al-Samadi et al (2013), where lesional biopsies from 5 Minor RAS subjects were compared with 5 controls, found that the staining of IL-8 and TNF-a were stronger in RAS lesions than controls. There were no significant differences reported in serum levels of IL-8 and TNF-a between the groups in our study (Al-Samadi et al 2013). However, these studies were limited by the lack of comparison to the Major or Herpetiform RAS, as well as by there not being enough participants in the control groups might affect the interpretation of their results.

6.4 Effects of different modalities of treatments on Cytokines:

Our study findings showed that treatment with Betamethasone mouthwash QDS and BD as a potent anti-inflammatory local steroid for 12 months, caused significant reductions of three salivary pro-inflammatory cytokines (IL- 181

6, IL-8, and TNF-a) and one anti-inflammatory cytokine (IL-10). Treatment with a low dose of systemic Colchicine 0.5mg a day for 12 months did not cause any significant changes in serum or salivary cytokines concentrations in spite of inducing significant improvements in the ulcer severity scores. However, combined treatment of Colchicine and Betnesol for 12 months, resulted in significant reductions of two salivary pro-inflammatory cytokines (IL-6 and IL-8), while in serum its effect was limited to the reduction of IL-8 only. No significant changes were demonstrated in the concentrations of IL-2, IL-4, IL-5 and IL-17 either in saliva or in serum.

Betamethasone in various forms and its effect on pro-inflammatory and anti- inflammatory cytokines were suggested in a few studies. Takahashi et al (1996), recruited 77 patients complained from sciatica pain following lumbar disc herniation. The study showed that IL-6 and TNF-a were significantly decreased with the use of a Betamethasone injection. Our study findings were comparable to Takahashi et al study, since our results showed that the local effect of Betamethasone mouthwash caused significant reductions in salivary concentrations of IL-6 and TNF-a (Takahashi et al 1996). However, Betamethasone mouthwash did not cause any significant changes in serum levels of IL-6 and TNF-a in our study.

In a randomized clinical trial of 80 women in premature labour, cervical fluids were collected before and 48 hours after treatment with Betamethasone. Endo-cervical concentration of inflammatory cytokines were analysed for the evaluation of IL-8, IL-17 and IFN-g, before and after treatment. The study showed that significant reductions of IL-8, IL-17 and IFN-g (p<0.05), were seen in the Betamethasone treated group (Khazardoust et al 2012). Our findings were in contrast to their results in some cytokines but not for others, since treatment with Betamethasone mouthwash did not cause any significant changes in serum levels of IL-8, IL-17 and IFN-g, though it caused significant reductions in salivary concentrations of IL-8. Nevertheless, combined treatment of Colchicine plus Betamethasone caused significant reductions in serum concentrations of IL-8.

182

Our study findings were in contrast to a study by Sakuma et al (2001), who used Tacrolimus ointment in comparison to Betamethasone valerate cream for the treatment of atopic dermatitis in animal models. Sakuma et al study showed that the suppressive effects of Tacrolimus on the productions of the cytokines involved (IL-2, IL-4, IL-5 and IFN-g), were equal to the effect of Betamethasone valerate cream (Sakuma et al 2001). Our results showed that treatment with Betamethasone mouthwash QDS and BD for 12 months did not cause any significant changes either in serum or in salivary concentrations of IL-2, IL-4, IL-5 and IFN-g,

Our study findings showed that combination therapy of Colchicine and Betnesol caused significant reductions in serum level of IL-8 as well as in salivary concentrations of IL-6 and IL-8. Similar results in some cytokines were reported in a study by Sun et al (2009), which included 43 patients with Behcet disease compared with 54 controls, all of whom were treated with Levamisole plus Colchicine for the periods of 0.5-11.5 months. Sun et al study showed that combined treatments with Levamisole and Colchicine caused significant reductions in serum concentrations of IL-6, IL-8 and TNF-a in 43 patients with Behcet disease. Although our findings and Sun et al results were in agreement regards IL-8, we have contradicted results regard IL-6 and TNF- a. This could be due to the action of systemic Levamisole. Further comparable results were reported in a study by Manie et al (1993), who demonstrated that Colchicine failed to affect the expression of TNF-a and IL-6 in human monocytes (Manie et al 1993).

In contrast to Sun et al findings and comparable to our results were demonstrated in a study by Celkan et al (2005). This study with a large sample size of 70 children with Familial Mediterranean Fever disease (FMF), the subjects were divided into three groups (group one; 17 patients newly diagnosed and untreated, group two; 36 patients on Colchicine therapy and group three; 17 healthy controls). The study showed that there were no significant changes in the serum levels of IL-6 between the first two groups. Although, the authors suggested that Colchicine had a positive effect on the disease activity, there was no correlation between disease activity and 183 cytokines concentrations (Celkan et al 2005). Our study findings were comparable with this study, since there were no significant changes in serum levels of IL-6, before and after treatment with Colchicine.

Our study findings appear to differ from those of Woo et al (2012). Their findings showed that Colchicine significantly modulated the concentrations of IL-6 and TNF-a in peripheral blood mononuclear cells (PBMC) from patients with Behcet’s disease on Colchicine therapy. However, the difference in the concentrations of these two pro-inflammatory cytokines was seen in the responder group and healthy controls but not in the non-responder group (Woo et al 2012). And it might well be that the changes seen in vitro with isolated mononuclear cells are different from those in serum reflecting in vivo activity. In addition, Miller et al (1992) examined Colchicine therapy and its effects on cytokines in the treatment of Primary Biliary Cirrhosis disease (PBC). The study included 28 patients with (PBC) who were tested before and during treatment with Colchicine compared with a placebo. The study noted a significant reduction in serum levels of IL-2 and TNF-a during Colchicine treatment while no changes were noticed in either IL-2 or TNF-a values in the placebo group (Miller et al 1992). Our study findings appear to contrast with their results, since our treatments with systemic Colchicine did not cause any significant changes either in serum or in salivary concentrations of IL-2, IL6 and TNF-a in RAS patients.

A benefit of salivary analysis of cytokines is that it is an easy and safe procedure, which permits the measurement of different cytokines at various periods in the same patient.

Finally, in order to determine whether levels of individual cytokines correlated with the Ulcer Severity Score (USS), all nine cytokines (IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-a and IFN-g) were measured before treatment (baseline visit) and their concentrations were correlated with the USS before treatment (baseline visit). No significant correlations were found. Furthermore, in order to determine whether changes in the concentrations of cytokines after treatment of 12 months with Betnesol mouthwash as monotherapy, systemic 184

Colchicine or Colchicine plus Betnesol mouthwash, were correlated with the changes in the USS after treatment, though there were no significant correlations found. Our study findings of no association between cytokines levels and disease activity or treatments, were in agreement with Wallace et al (2005) and Curnow et al (2008). Both these studies reported that the raised levels of cytokines did not correlate with the disease activity or treatments.

In summary, our study findings suggest that four salivary pro-inflammatory cytokines (IL-6, IL-8, TNF-a and IFN-g) and one anti-inflammatory cytokine (IL-10) are highly related to RAS. In addition, salivary IL-6 and IFN-g are directly related to the inflammatory process of RAS and to the severity of the disease. In serum, IL-8 is the only cytokine that was found to be significantly raised in RAS patients. Furthermore, this study showed that the anti- inflammatory effect of local steroid (Betamethasone mouthwash) modulated salivary levels of IL-6, IL-8, IL-10 and TNF-a, while combined therapy of systemic Colchicine and local steroid mouthwash modulated serum and salivary levels of IL-8 as well as salivary IL-6.

185

CONCLUSIONS OF THE STUDY

The results of this study suggest:

1. Nine months treatment with Betnesol mouthwash QDS during ulcer attacks and BD in between could be used as a first line standard and safe therapy for the management of Minor RAS. It was effective in reductions of size, number, duration, pain and site and has the potential to prevent recurrences during the treatment period with no side effects reported.

2. Nine months treatment with Colchicine 0.5mg OD and Betnesol mouthwash QDS during ulcer attacks for Major RAS. This combined therapy was effective in reductions of size, number, duration, pain and site and has the potential to prevent recurrences during the treatment period with few adverse events reported. In addition, we suggest that 6 months treatment with 1mg of Colchicine OD was more effective for recalcitrant Major RAS patients.

3. Five salivary cytokines (IL-6, IL-8, IL-10 TNF-a and IFN-g) are highly related to RAS. In addition, salivary IL-6 and IFN-g are directly related to the inflammatory process of RAS and to the severity of the disease. In serum, IL- 8 is the only cytokine that was found significantly higher in RAS.

4. Betnesol mouthwash can modulate salivary levels of IL-6, IL-8, IL-10 and TNF-a, while combined therapy of systemic Colchicine and Betnesol mouthwash can modulate serum and salivary levels of IL-8 as well as salivary IL-6 in RAS patients.

186

FUTURE WORK

There is a need to extend this study to a larger patient group which includes a statistically significant number of Herpetifrom RAS subjects. The efficacy of Betnesol mouthwash with other systemic treatment or with a higher dose of Colchicine needs to be compared in a multi-centre, double-blind and placebo- controlled study. Further studies are needed to determine the normal range of different salivary cytokines in Behcet disease and RAS with the effect of different local and systemic medications on these cytokines.

187

References

Akdeniz N, Esrefoglu M, Keles MS, Karakuzu A, Atasoy M. (2004) Serum Interleukin-2, Interleukin-6, tumour necrosis factor-Alpha and nitric oxide levels in patients with Behcet's disease. Annals-Academy of Medicine Singapore. 33(5):596-9.

Akman A, Ekinci NC, Kacaroglu H, Yavuzer U, Alpsoy E, Yegin O. (2008) Relationship between periodontal findings and specific polymorphisms of interleukin-1alpha and -1beta in Turkish patients with Behçet's disease. Arch Dermatol Res. 300(1):19-26.

Akman A, Kacaroglu H, Donmez L, Bacanli A, Alpsoy E. (2007) Relationship between periodontal findings and Behçet's disease: a controlled study. J Clin Periodontol. 34(6):485-91.

Albanidou-Farmaki E, Kayavis IG, Polymenidis Z, Papanayotou P. (1988) HLA-A, B, C and DR antigens in recurrent oral ulcers. Ann Dent. 47:5-8.

Albanidou-Farmaki E, Markopoulos AK, Kalogerakou F, Antoniades DZ. (2007) Detection, enumeration and characterization of T helper cells secreting type 1 and type 2 cytokines in patients with recurrent aphthous stomatitis. Tohoku J Exp Med. 212(2):101-5.

Al-Na'mah ZM, Carson R, Thanoon IA. (2009) Dexamucobase: a novel treatment for oral aphthous ulceration. Quintessence Int. 40(5):399-404.

Al-Omiri MK, Karasneh J, Lynch E. (2012) Psychological profiles in patients with recurrent aphthous ulcers. Int J Oral Maxillofac Surg. 41(3):384-8.

Alsahaf S, Tappuni AR, Challacombe SJ. (2012) Management of recurrent aphthous stomatitis with colchicine and betnesol mouthwash. Oral Diseases. 18: 30-30.

Alsahaf S, Tappuni A, McParland H. (2010) Evaluation of the efficacy of Colchicine with and without Betnesol mouthwash in management of RAS . Oral Diseases. 16(6):546-546.

Alsahaf S, Cook RJ, Matharu H. (2010) Oral Disease Severity Scores (ODSS) and Treatment in patients with Recurrent Ulcers. Oral Diseases. 16(6):546-546.

188

Al-Samadi A, Kouri VP, Salem A, Ainola M, Kaivosoja E, Barreto G, Konttinen YT, Hietanen J, Hayrinen-Immonen R. (2014) IL-17C and its receptor IL-17RA/IL-17RE identify human oral epithelial cell as an inflammatory cell in recurrent aphthous ulcer. J Oral Pathol Med. 43(2):117-24.

Andrews VH, Hall HR. (1990) The effect of relaxation/imagery training on recurrent aphthous stomatitis: a preliminary study. Psychosom Med. 52;526-535.

Axell T, Henricsson V. (1985) The occurrence of recurrent aphthous ulcers in an adult Swedish population. Acta Odontol Scand. 43;121-125.

Babaee N, Zabihi E, Mohseni S, Moghadamnia AA. (2012) Evaluation of the therapeutic effects of Aloe vera gel on minor recurrent aphthous stomatitis. Dent Res J (Isfahan). 9(4):381-5.

Babaee N, Mansourian A, Momen-Heravi F, Moghadamnia A, Momen- Beitollahi J. (2010) The efficacy of a paste containing Myrtus communis (Myrtle) in the management of recurrent aphthous stomatitis: a randomized controlled trial. Clin Oral Investig. 14(1):65-70.

Baccaglini L, Theriaque DW, Shuster JJ, Serrano G, Lalla RV. (2013) Validation of anamnestic diagnostic criteria for recurrent aphthous stomatitis. J Oral Pathol Med. 42(4):290-4.

Bachtiar EW, Cornain S, Siregar B, Raharjo TW. (1998) Decreased CD4+/CD8+ ratio in major type of recurrent aphthous ulcers: comparing major to minor types of ulcers. Asian Pac J Allergy Immunol. 16(2-3):75-9.

Bagán J.V, J. M. Sanchis, M. A. Milián, M. Peñarrocha, F. J. Silvestre. (1991) Recurrent aphthous stomatitis. A study of the clinical characteristics of lesions in 93 cases. Journal of Oral Pathology & Medicine. 20(8):395–397.

Balan U, Gonsalves N, Jose M, Girish KL. (2012) Symptomatic changes of oral mucosa during normal hormonal turnover in healthy young menstruating women. J Contemp Dent Pract. 1;13(2):178-81.

Bang D, Hur W, Lee ES, Lee S. (1995) Prognosis and clinical relevance of recurrent oral ulceration in Behçet's disease. The Journal of Dermatology. 22(12):926-929.

189

Baroody Fuad M, Rouadi P, Driscol Peter V, Bochner Bruce S, Naclerio Robert M. (1998) Intranasal Beclomethasone Reduces Allergen-induced Symptoms and Superficial Mucosal Eosinophilia without Affecting Submucosal Inflammation. American J of Respiratory and Crit. Care Med. 157(3):899-906.

Barton, BE, Jakway JP, Smith SR, Siegel MI. (1991) Cytokine Inhibition by A Novel Steroid, Mometasone Furoate Immunopharmacology and Immunotoxicology. 13(3):251 – 261.

Bazrafshani MR, Hajeer AH, Ollier WE, Thornhill MH. (2002) Recurrent aphthous stomatitis and gene polymorphisms for the inflammatory markers TNF-alpha, TNF-beta and the vitamin D receptor: no association detedted. Oral Dis. 8(6):303-7.

Bazrafshani MR, Hajeer AH, Ollier WE, Thornhill MH. (2002) IL-1B and IL-6 gene polymorphisms encode significant risk for the development of recurrent aphthous stomatitis (RAS). Genes Immun. 3(5):302-5.

Bell J. (2005) Amlexanox for the treatment of recurrent aphthous ulcers. Clin Drug Investig. 25(9):555-66.

Ben-Chetrit E, S. Bergmann, R. Sood. (2006) Mechanism of the anti-inflammatory effect of colchicine in rheumatic diseases: a possible new outlook through microarray analysis. Rheumatology. 45(3):274-282.

Bhatnagar P, Rai S, Bhatnagar G, Kaur M, Goel S, Prabhat M. (2013) Prevalence study of oral mucosal lesions, mucosal variants, and treatment required for patients reporting to a dental school in North India: In accordance with WHO guidelines. J Family Community Med. 20(1):41-8.

Boras VV, Lukac J, Brailo V, Picek P, Kordi ć D, Zili ć IA. (2006) Salivary interleukin-6 and tumor necrosis factor-alpha in patients with recurrent aphthous ulceration. J Oral Pathol Med. 35(4):241-3.

Brown RS, Bottomley WK. (1990) Combination an immunosuppressant and topical steroids therapy for treatment of recurrent major aphthae. Oral Surg Oral Med Oral Pathol. 69(1):42-4.

Brozovic S, Vucicevic-Boras V, Mravak-Stipetic M, Jukic S, Kleinheinz J, Lukac. (2002) Salivary levels of vascular endothelial growth factor (VEGF) in recurrent aphthous ulceration.

190

J Oral Pathol Med. 31(2):106-8.

Buño IJ, Huff JC, Weston WL, Cook DT, Brice SL. (1998) Elevated levels of interferon gamma, tumor necrosis factor alpha, interleukins 2, 4, and 5, but not interleukin 10, are present in recurrent aphthous stomatitis. Arch Dermatol. 134(7):827-31.

Celkan T, Celik M, Kasap Copur O, Ozkan A, Apak H, Ocak S, Arisoy N, Yildi I. (2005) The Anaemia of Familial Mediterranean Fever Disease. Pediatric Hematology and Oncology. 22(8):657–665.

Challacombe SJ, Barkhan P, Lehner T. (1977) Haematological features and differentiation of recurrent oral ulceration. Brit. J. Oral Surg. 15:37-48.

Challacombe SJ, Batchelor JR, Kennedy LA, Lehner T. (1977) HLA antigens in recurrent oral ulcerations. Arch Dermatol. 113;1717-1719.

Challacombe SJ, Shirlaw PJ. (1991) Oral ulceration: when to treat, refer or ignore. Dent Update. 18(9):368-73.

Chi W, Yang P, Zhu X, Wang Y, Chen L, Huang X, Liu X. (2010) Production of interleukin-17 in Behcet's disease is inhibited by cyclosporin A. Mol Vis. 19(16):880-6.

Chang HK, Kim JU, Cheon KS, Chung HR, Lee KW, Lee IH. (2001) HLA-B51 and its allelic types in association with Behcet's disease and recurrent aphthous stomatitis in Korea. Clin Exp Rheumatol. 19(5 Suppl 24):S31-5.

Choi SSL, Chan KF, Ng HK, Mak WP. (1999) Colchicine-induced myopathy and neuropathy. HKMJ. 5:204-7.

Cooke BE. (1969) Recurrent oral ulceration. Br J Dermatol. 81(2):159-61.

Corbel M, Germain N, Lanchou J, Molet S, R e Silva MR, Martins MA, Boichot E, Lagente V. (2002) The Selective Phosphodiesterase 4 Inhibitor RP 73-401 Reduced Matrix Metalloproteinase 9 Activity and Transforming Growth Factor-Release During Acute Lung Injury in Mice: The Role of the Balance Between Tumor Necrosis Factor-alpha and Interleukin-10. Journal of Pharmacology and Experimental Therapeutics. 301(1):258-265.

191

Crivelli MR, Aguas S, Adler I, Quarracino C, Bazerque P. (1988) Influence of cocioeconomic status on oral mucosal lesion prevalence in school children. Community Dent and Oral Epidemiol. 16(1):58-60.

Cronstein BN, Molad Y, Reibman J, Balakhane E, Levin RI, Weissmann G. (1995) Colchicine alters the quantitative and qualitative display of selectins on endothelial cells and neutrophils. J Clin Invest. 96(2):994-1002.

Curnow SJ, Pryce K, Modi N, Knight B, Graham EM, Stewart JE, Fortune F, Stanford MR, Murray PI, Wallace GR. (2008) Serum cytokine profiles in Behçet's disease: is there a role for IL-15 in pathogenesis? Immunol Lett. 16;121(1):7-12.

Dalghous AM, Freysdottir J, Fortune F. (2006) Expression of cytokines, chemokines, and chemokine receptors in oral ulcers of patients with Behcet's disease (BD) and recurrent aphthous stomatitis is Th1-associated, although Th2-association is also observed in patients with BD. Scand J Rheumatol. 35(6):472-5.

De Abreu MA, Hirata CH, Pimentel DR, Weckx LL. (2009) Treatment of recurrent aphthous stomatitis with clofazimine. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 108(5):714-21.

Direskeneli H. (2001) Behçet's disease: infectious aetiology, new autoantigens, and HLA-B51 Ann Rheum Dis. 60:996-1002.

Dolby AE, Walker DM, Slade M, Allan C. (1977) HLA histocombatibility antigens in recurrent aphthous ulceration. J Dent Res. 56;105-107.

Edres MA, Scully C, Gelbier M. (1997) Use of proprietary agents to relieve recurrent aphthous stomatitis. Br Dent J. 22;182(4):144-6.

Eguia-del Valle A, Martinez-Conde-Lamosas R, Lopez-Vicente J, Uribarri- Etxebarria A, Aguirre-Urizar JM. (2011) Salivary levels of Tumour Necrosis Factor-alpha in patients with recurrent aphthous stomatitis. Med Oral Patol Oral Cir Bucal. 1;16(1):e33-6.

Eguia-del Valle A, Martínez-Conde-Llamosas R, López-Vicente J, Uribarri- Etxebarria A, Aguirre-Urizar JM. (2013) Salivary cortisol determination in patients from the Basque Country with recurrent aphthous stomatitis. A pilot study

192

Med Oral Patol Oral Cir Bucal. 1;18(2):e207-11.

Embil JA, Stephens RG, Munuel FR. (1975) Prevelance of recurrent herpes labialis and aphthous ulcers among young adults on six continents. CMA Journal. 113:627-630.

Edward A, Graykowski. (1978) Clinical evaluation of benzydamine, chlorhexidine, and placebo mouthwashes in the management of recurrent aphthous stomatitis. Journal of Oral Pathology & Medicine. 6(7):376-382-.

Fabiana M. Barros, Monica A. Lotufo, Priscila M. Andrade, Cristiane M. Franca, and Ricardo C. Borra. (2010) Possible Association between Th1 Immune Polarization and Epithelial Permeability with Toll-Like Receptors 2 Dysfunction in the Pathogenesis of the Recurrent Aphthous Ulceration. Leopoldo Mandic Dental Research Institute. Article ID 163804.

Fahmy MS. (1976) Recurrent aphthous ulcers in a mixed Arab community. Community Dent Oral Epidemiol. 4;160-164.

Fani MM, Ebrahimi H, Pourshahidi S, Aflaki E, Shafiee Sarvestani S. (2012) Comparing the Effect of Phenytoin Syrup and Triamcinolone Acetonide Ointment on Aphthous Ulcers in Patients with Behcet's Syndrome. Iran Red Crescent Med J. 14(2):75-8.

Femiano F, Buonaiuto C, Gombos F, Lanza A, Cirillo N. (2010) Pilot study on recurrent aphthous stomatitis (RAS): a randomized placebo- controlled trial for the comparative therapeutic effects of systemic prednisone and systemic montelukast in subjects unresponsive to topical therapy. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 109(3):402-7.

Ferguson MM, Carter J, Boyle P. (1984) An epidemiological study of factors associated with recurrent aphthous in women. J Oral Med. 39:212-217.

Field EA, Brookes V, Tyldesley WR. (1992) Recurrent Aphthous ulceration in children; a review. Int J Pediatric Dent. 2;1-10.

Fontes V, Machet L, Huttenberger B, Lorette G, Vaillant L. (2002) Recurrent aphthous stomatitis: treatment with colchicine. An open trial of 54 cases. Ann Dermatol Venereol. 129(12):1365-9.

193

FreysdottirJ, Lau SH, Fortune F. (1999) Gammadelts T cells in Behcets disease (BD) and recurrent aphthous stomatitis (RAS). Clinc Exp Immunol. 118(3):451-7.

Gallina G, Cumbo V, Messina P, Caruso C. (1985) HLA-A, B, C, DR, MT, MB antigen in recurrent aphthous stomatitis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 59:364-370.

Gallo Cde B, Mimura MA, Sugaya NN. (2009) Psychological stress and recurrent aphthous stomatitis. Clinics (Sao Paulo). 64(7):645-8.

Genvo MF, Faure M, Thivolet J. (1984) Treatment of aphthosis with thalidomide and with colchicine. Dermatologica. 168(4):182-8.

Gil H, Perrin S, Dupond JL, Meaux-Ruault N, Hafsaoui C, Limat S, Magy- Bertrand N. (2010) Recurrent aphthosis: safety of low dose thalidomide. Rev Med Interne. 31(6):403-5.

Gorsky M, Epstein J, Rabenstein S, Elishoov H, Yarom N. (2007) Topical minocycline and tetracycline rinses in treatment of recurrent aphthous stomatitis: a randomized cross-over study. Dermatol Online J. 1;13(2):1.

Gorsky M, Epstein J, Raviv A, Yaniv R, Truelove E. (2008) Topical minocycline for managing symptoms of recurrent aphthous stomatitis. Spec Care Dentist. 28(1):27-31.

Górska R. (1997) Epidemiologic studies of oral mucosa changes occurring in children, adolescents, and adults 13-24 years of age in Warsaw. Przegl Epidemiol. 51(3):339-47.

Graykowski EA, Kingman A. (1978) Double-blind trial of tetracycline in recurrent aphthous ulcerations. J Oral Pathol. 7:376-382.

Greenspan JS, Gadol N, Olson JA. (1985) Lymphocytes function in recurrent aphthous ulceration. J Oral Pathology. 14:592-6.

Günzler V. (1992) Thalidomide in human immunodeficiency virus (HIV) patients. A review of safety considerations. Drug Saf. 7(2):116-34.

194

Guimarães AL, de Sá AR, Victória JM, Correia-Silva JF, Pessoa PS, Diniz MG, Gomez RS. (2006) Association of interleukin-1beta polymorphism with recurrent aphthous stomatitis in Brazilian individuals. Oral Dis.J. 12(6):580-3.

Habich C, Burkat V. (2007) Heat shock protein 60: regulatory role on innate immune cells. Cell Mol Life Sci. 64(6):742-51.

Handa R, Bailoor DN, Desai VD, Sheikh S, Goyal G. (2012) A study to evaluate the impact of examination stress on recurrent aphthous ulceration in professional college students in Jaipur district. Minerva Stomatol. 61(11-12):499-507.

Handfield-Jones S, Allen BR, Littlewood SM. (1985) Dapsone use in oral genital ulcers. Br J Dermatol. 113:501.

Hansten PD, Horn JR. (2011) Cytochrome P450 Enzymes and Transporter Table” in, The Top 100 Drug Interactions. A guide to patient management. H&H Publications, Edmonds, WA, 2011.

Hapa A, Aksoy B, Polat M, Aslan U, Atakan N. (2011) Does recurrent aphthous stomatitis affect quality of life? A prospective study with 128 patients evaluating different treatment modalities. J Dermatolog Treat. 22(4):215-20.

Hasan A, Shinnick T, Mizushema V, Van Der Zee R , Lehner T. (2002) Defining a T-cell epitope within HSP 65 in recurrent aphthous stomatitis. Clinc Exp Immuno. 128(2):318-25.

Hasan A, Childerstone A, Pervin K, Shinnick T, Mizushima Y, Van Der Zee R, Vaughan R, Lehner T. (1995) Recognition of a unique peptide epitope of the mycobacterial and human heat shock protein 65-60 antigen by T cells of patients with recurrent oral ulcers. Clin Exp Immunol. 99(3):392-397.

Hassan A, Z Ahmed, R Ali, F Ara, N Ahmed, MU Salam. (2010) Colchicine in the Treatment of Recurrent Oral Aphthous Ulcer-Open Clinical Trial in Bangladesh. Bangladesh Medical Journal. 3(39).

Harms M. (1980) Alopecia and hair changes following colchicine therapy. Hautarzt. 31(3):161-3.

195

Hegarty AM, Hodgson TA, Lewsey JD, Porter SR. (2002) Fluticasone propionate spray and betamethasone sodium phosphate mouthrinse: a randomized crossover study for the treatment of symptomatic oral lichen planus. J Am Acad Dermatol. 47(2):271-9.

Hello M, Barbarot S, Bastuji-Garin S, Revuz J, Chosidow O. (2010) Use of thalidomide for severe recurrent aphthous stomatitis: a multicenter cohort analysis. Medicine (Baltimore). 89(3):176-82.

Henricsson V, Axéll T. (1985) Treatment of recurrent aphthous ulcers with Aureomycin mouth rinse or Zendium dentifrice. Acta Odontol Scand. 43(1):47-52.

Huling LB, Baccaglini L, Choquette L, Feinn RS, Lalla RV. (2012) Effect of stressful life events on the onset and duration of recurrent aphthous stomatitis. J Oral Pathol Med. 41(2):149-52.

Hunter L, Addy M. (1987) Chlorhexidine gluconate mouthwash in the management of minor aphthous ulceration. A double-blind, placebo-controlled cross-over trial. Br Dent J. 7;162(3):106-10.

Irakam A, Miskolci V, Vancurova I, Davidson D. (2002) Dose-Related Inhibition of Proinflammatory Cytokine Release from Neutrophils of the Newborn by Dexamethasone, Betamethasone, and Hydrocortisone. Biol Neonate. 82:89-95.

Jaber L, Weinberger A, Klein T, Yaniv I, Mukamel M. (2001) Close association of HLA-B52 and HLA-B44 antigens in Israeli Arab adolescents with recurrent aphthous stomatitis. Arch Otolaryngol Head Neck Surg. 127(2):184-7.

Jiang XW, Zhang Y, Zhu YL, Zhang H, Lu K, Li FF, Peng HY. (2013) Effects of berberine gelatin on recurrent aphthous stomatitis: a randomized, placebo-controlled, double-blind trial in a Chinese cohort. Oral Surg Oral Med Oral Pathol Oral Radiol. 115(2):212-7.

Jorizzo JL, Taylor RS, Schmalstieg FC, Solomon AR JR, Daniels JC, Rudloff HE, Cavallo T. (1985) Complex aphthosis: a forme fruste of Behçet's syndrome? J Am Acad Dermatol. 13(1):80-4.

Jurge, Kuffer, Scully C, Porter S. (2006) Mucosal disease series . Oral Diseases. 12(1):1-21.

196

Kalkan G, Karakus N, Yigit S. (2014) Association of MTHFR gene C677T mutation with recurrent aphthous stomatitis and number of oral ulcers. Clin Oral Investig. 18(2):437-41.

Kalkan G, Yigit S, Karakus N, Ba ş Y, Pancar GŞ, Balta I. (2013) Association between MEFV gene mutations and recurrent aphthous stomatitis in a cohort of Turkish patients. J Dermatol. 40(7):516-21.

Karakus N, Yigit S, Rustemoglu A, Kalkan G, Bozkurt N. (2014) Effects of interleukin (IL)-6 gene polymorphisms on recurrent aphthous stomatitis. Arch Dermatol Res. 306(2):173-80.

Karavana SY, Gökçe EH, Rençber S, Özbal S, Pekçetin C, Güneri P, Ertan G. (2012) A new approach to the treatment of recurrent aphthous stomatitis with bioadhesive gels containing cyclosporine A solid lipid nanoparticles: in vivo/in vitro examinations. Int J Nanomedicine. 7:5693-704.

Katz J, Langevitz P, Shemer J, Barak S, Livneh A. (1994) Prevention of recurrent aphthous stomatitis with colchicine: an open trial. J Am Acad Dermatol. 31(3 Pt 1):459-61.

Keenan AV, Spivakovksy S. (2013) Stress associated with onset of recurrent aphthous stomatitis. Evid Based Dent. 14(1):25.

Khanzardoust S, Javadian P, Salmanian B, Zandevakil F, Abbasalizadeh F, Alimonhamadi S, Borna S, Ghazanfari T, Hantoushzadeh S. (2012) A clinical randomized trial on endocervical inflammatory cytokines and betamethasone in prime-gravid pregnant women at risk of preterm labor. Iran J Immunol. 9(3):199-207.

Kim Y, Greenberg MS. (2001) Management of patients with severe oral mucosal disease. Alpha Omegan. 94(2):18-23.

Kiraz S, Ertenli I, Arici M, Calgüneri M, Haznedaroglu I, Celik I, Pay S, Kirazli S. (1998) Effects of colchicine on inflammatory cytokines and selectins in familial Mediterranean fever. Clin Exp Rheumatol. 16(6):721-4.

Konda C, Rao AG. (2010) Colchicine in dermatology. Indian J Dermatol Venereol Leprol. 76(2):201-5.

197

Koray M, AK G, Kurklu E, Tanyeri H, Ayden F, Ogus FS, Temurhan S, Ciltci H, Carin M, Onal AE, Ozdilli A. (2009) The effect of glucan on recurrent aphthous stomatitis. The Journal Of Alternative And Complementory Medicine. 15(2):111-3.

Koridze Kh, Ladashvili L, Taboridze I, Bakradze M. (2007) Immunological aspects of aphthous stomatitis. Georgian Med News. 151:37-9.

Krisdapong S, Sheiham A, Tsakos G. (2012) Impacts of recurrent aphthous stomatitis on quality of life of 12- and 15-year- old Thai children. Qual Life Res. 21(1):71-6.

Lalla RV, Choquette LE, Feinn RS, Zawistowski H, Latortue MC, Kelly ET, Baccaglini L. (2012) Multivitamin therapy for recurrent aphthous stomatitis: a randomized, double- masked, placebo-controlled trial. J Am Dent Assoc. 143(4):370-6.

Lehner T, Poulter LW. (1989) Immunohistology of oral lesions from patients with recurrent oral ulcers and Behçet's syndrome. Clin Exp Immunol. 78(2):189-95.

Lehner T. (1968) Autoimmunity in oral diseases, with special reference to recurrent oral ulceration. Proceedings of the Royal Society of Medicine. 61(5):515-524.

Lehner T. (1978) Immunological aspects of recurrent oral ulceration and Behcets syndrome. J of oral pathol. 7(6):424-430.

Lehner T, Batchelor JR, Challacombe SJ, Kennedy L. (1979) An Immunogenetic basis for the tissue involvement in Behcets Syndrome. Immunology Letters. 37(4):895-900.

Lehner T, Lavery E, Smith R, van der Zee R, Mizushima Y, Shinnick T. (1991) Association between the 65-kilodalton heat shock protein, Streptococcus sanguis, and the corresponding antibodies in Behçet's syndrome. Infect Immun. 59(4):1434-41.

Lehner T, Wilton JM, Ivanyi L. (1976) Double blind crossover trial of levamisole in recurrent aphthous ulceration. Lancet. 30;2(7992):926-9.

Lehner T. (1977) Oral ulceration and Behcet's syndrome. Gut. 18(6):491-511. Review.

198

Letsinger JA, McCarty MA, Jorizzo JL. (2005) Complex aphthosis: a large case series with evaluation algorithm and therapeutic ladder from topicals to thalidomide. J Am Acad Dermatol. 52(3 Pt 1):500-8.

Lewkowicz N, Lewkowicz P, Banasik M, Kurnatowska A, Techorzewski H. (2005) Predominance of type 1 cytokines and decreased number of CD4 + CD25 +high T regulatory cells in peripheral blood of patients with recurrent aphthous ulcerations. Immunol letters. 15;99(1):57-62.

Lewkowicz N, Kur B, Kurnatowska A, Tchorzewski H, Lewkowicz P. (2011) Expression of Th1/Th2/Th3/Th17-related genes in recurrent aphthous ulcers. Arch Immunol Ther Exp (Warsz). 59(5):399-406.

Liang MW, Neoh CY. (2012) Oral aphthosis: Management gaps and recent advances. Ann Acad Med Singapore. 41(10):463-70.

Liu J, Zeng X, Chen Q, Cai Y, Chen F, Wang Y, Zhou H, Lin M, Shi J, Wang Z, Zhang Y. (2006) An evaluation on the efficacy and safety of amlexanox oral adhesive tablets in the treatment of recurrent minor aphthous ulceration in a Chinese cohort: a randomized, double-blind, vehicle-controlled, unparallel multicenter clinical trial. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 102(4):475-81.

Liu C, Zhou Z, Liu G, Wang Q, Chen J, Wang L, Zhou Y, Dong G, Xu X, Wang Y, Gou Y, Lin M, Wu L, Du G, Wei C, Zeng X, Wang X, Wu J, Li B, Zhou H. (2012) Efficacy and safety of dexamethasone ointment on recurrent aphthous ulceration. Am J Med. 125(3):292-301.

Liu X, Guan X, Chen R, Hua H, Liu Y, Yan Z. (2012) Repurposing of yunnan baiyao as an alternative therapy for minor recurrent aphthous stomatitis. Evid Based Complement Alternat Med. 2012:284620.

Lynde CB, Bruce AJ, Rogers RS 3rd. (2009) Successful treatment of complex aphthosis with colchicine and dapsone. Arch Dermatol. 145(3):273-6.

Matthews RW, Scully CM, Levers BG, Hislop WS. (1987) Clinical evaluation of benzydamine, chlorhexidine, and placebo mouthwashes in the management of recurrent aphthous stomatitis. Oral Surg Oral Med Oral Pathol. 63(2):189-91.

199

Manie S, Schmid-Alliana A, Kubar J, Ferrua B, Rossi B. (1993) Disruption of microtubule network in human monocytes induces expression of interleukin-1 but not that of interleukin-6 nor tumor necrosis factor-alpha. Involvement of protein kinase A stimulation. J. Biol. Chem. 25;268(18):13675-81.

Majorana A, Bardellini E, Flocchini P, Amadori F, Conti G, Campus G. (2010) Oral mucosal lesions in children from 0 to 12 years old: ten years' experience. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 110(1):13-8.

McCartan BE, Sullivan A. (1992) The association of menstrual cycle, pregnancy, and menopause with recurrent oral aphthous stomatitis: a review and critique. Obstet Gynecol. 80(3 Pt 1):455-8. Review.

Mekori Y A, Chowers Y, Ducker I, Klajman A. (1989) Inhibition of delayed hypersensitivity reactions by colchicine. II. Colchicine inhibits interferon-gamma induced expression of HLA-DR on gut epithelial cell line. Clin Exp Immunol. 78(2): 230–232.

Mege JL, Dilsen N, Sanguedolce V, Gul A, Bongrand P, Roux H, Ocal L, Inanç M, Capo C. (1993) Over production of monocyte derived tumor necrosis factor alpha, interleukin (IL) 6, IL-8 and increased neutrophil superoxide generation in Behçet's disease. J Rheumatol. 20(9):1544-9.

Meng W, Dong Y, Liu J, Wang Z, Zhong X, Chen R, Zhou H, Lin M, Jiang L, Gao F, Xu T, Chen Q, Zeng X. (2009) A clinical evaluation of amlexanox oral adhesive pellicles in the treatment of recurrent aphthous stomatitis and comparison with amlexanox oral tablets: a randomized, placebo controlled, blinded, multicenter clinical trial. Trials. 6;10:30.

Merchant HW, Gangarosa LP, Glassman AB, Sobel RE. (1978) Betamethasone-17-benzoate in the treatment of recurrent aphthous ulcers. Oral Surg Oral Med Oral Pathol. 45(6):870-5.

Miles DA, Bricker SL, Razmus TF, Potter RH. (1993) Triamcinolone acetonide versus chlorhexidine for treatment of recurrent stomatitis. Oral Surg Oral Med Oral Pathol. 75(3):397-402.

Miller LC, Kaplan MM. (1992) Serum interleukin-2 and tumor necrosis factor-alpha in primary biliary cirrhosis: decrease by colchicine and relationship to HLA-DR4. Am J Gastroenterol. 87(4):465-70.

200

Miller MF, Ship II. A retrospective study of the prevalence and incidence of recurrent aphthous ulcers in a professional population, 1958-1971. Oral Surg Oral Med Oral Pathol. 1977;43(4):532-7.

Miller MF, Garfunkel AA, Ram CA, Ship II. (1980) The inheritance of recurrent aphthous stomatitis. Observations on susceptibility. Oral Surg Oral Med Oral Pathol. 49(5):409-12.

Miller MF, Garfunkel AA, Ram C, Ship II. (1977) Inheritance patterns in recurrent aphthous ulcers: twin and pedigree data. Oral Surg Oral Med Oral Pathol. 43(6):886-91.

Mills MP, Mackler BF, Nelms DC, Peavy DL. (1980) Quantitative distribution of inflammatory cells in recurrent aphthous stomatitis. J Dent Res. 59(3):562-6.

Mimura MA, Hirota SK, Sugaya NN, Sanches Jr JA, Migliari DA. (2009) Systemic treatment in severe cases of recurrent aphthous stomatitis: an open trial. Clinics (Sao Paulo). 64(3):193-8.

Misumi M, Hagiwara E, Takeno M, Takeda Y, Inoue Y, Tsuji T, Ueda A, Nakamura S, Ohno S, Ishigatsubo Y. (2003) Cytokine production profile in patients with Behcet's disease treated with infliximab. Cytokine. 7;24(5):210-8.

Miyachi Y, Taniguchi S, Ozaki M, Horio T. (1981) Colchicine in the treatment of the cutaneous manifestations of Behçet's disease. Br J Dermatol. 104(1):67-9.

Mumcu G, Sur H, Inanc N, Karacayli U, Cimilli H, Sisman N, Regun T, Direskeneli H. (2009) A composite index for determining the impact of oral ulcer activity in Behcet's disease and recurrent aphthous stomatitis. J Oral Pathol Med. 38(10):785-91.

Natah SS, Häyrinen-Immonen R, Hietanen J, Malmström M, Konttinen YT. (2000) Immunolocalization of tumor necrosis factor-alpha expressing cells in recurrent aphthous ulcer lesions (RAU). J Oral Pathol Med. 29(1):19-25.

Natah SS, Konttinen YT, Enattah NS, Ashammakhi N. (2004) Recurrent aphthous ulcers today: a review of the growing knowledge. Int J Oral Maxillofac Surg. 33(3);221-234.

201

O'Duffy JD, Carney JA, Deodhar S. (1971) Behçet's disease. Report of 10 cases, 3 with new manifestations. Ann Intern Med. 75(4):561-70.

O'Duffy JD, Taswell HF, Elveback LR.(1976) HL-A antigens in Behcet's disease. J Rheumatol. 3(1):1-3.

Ono S, Nakayama E, Sugiura S, Itakura K, Aoki K. (1975) Specific histocompatibility antigens associated with Behcet's disease. American Journal of Ophthalmology. 80(4):636-641

Olson JA, Feinberg I, Silverman S Jr, Abrams D, Greenspan JS. (1982) Serum vitamin B12, folate, and iron levels in recurrent aphthous ulceration. Oral Surg Oral Med Oral Pathol. 54(5):517-20.

Ozbakir F, Yazici H, Mat C, Tuzun Y, Yurkakul S, Yilmazer S. (1987) HLA antigens in recurrent oral ulceration: evidence against a common disease spectrum with Behcet’s syndrome. Clin Exp Reheumatol. 5:263-265.

Oztas MO, Onder M, Gurer M A, Bukan N, Sancak B. (2005) Serum interleukin 18 and tumour necrosis factor-α levels are increased in Behcet's disease. Clinical and Experimental Dermatology. 30(1):61-3.

Pakfetrat A, Mansourian A, Momen-Heravi F, Delavarian Z, Momen-Beitollahi J, Khalilzadeh O, Basir-Shabestari S. (2010) Comparison of colchicine versus prednisolone in recurrent aphthous stomatitis: A double-blind randomized clinical trial. Clin Invest Med. 1;33(3):189-95.

Pedersen A. (1989) Psychologic stress and recurrent aphthous ulceration. J Oral Pathol Med. 18(2):119-22.

Pekiner FN, Aytugar E, Demirel GY, Borahan MO. (2012) Interleukin-2, interleukin-6 and T regulatory cells in peripheral blood of patients with Behçet's disease and recurrent aphthous ulcerations. J Oral Pathol Med. 41(1):73-9.

Picciani BL, Silva-Junior GO, Barbirato DS, Ramos RT, Cantisano MH. (2010) Regression of major recurrent aphthous ulcerations using a combination of intralesional corticosteroids and levamisole: a case report. Clinics (Sao Paulo). 65(6):650-2.

Picek P, Buljan D, Rogulj AA, Stipeti ć-Ovcari ćek J, Cati ć A, Plestina S, Boras VV, Vidovi ć-Juras D. (2012) Psychological status and recurrent aphthous ulceration. Coll Antropol. 36(1):157-9.

202

Platz P, Ryder LP, Donatsky O. (1976) No deviations of HLA-A and -B antigens in patients with recurrent aphthous stomatitis. Tissue Antigens. 8(4):279-80.

Poulter LW, Lehner T. (1989) Immunohistology of oral lesions from patients with recurrent oral ulcers and Behcets syndrome. Clin Exp Immunol. 78(2):189-195.

Prasad RS, Pai A. (2013) Assessment of immediate pain relief with laser treatment in recurrent aphthous stomatitis. Oral Surg Oral Med Oral Pathol Oral Radiol. 116(2):189-93.

Rasool S, Ali A, Bashir F. (2007) Oral lichen planus. J Coll Physicians Surg Pak. 17(12):764-5.

Rees TD, Binnie WH. (1996) Recurrent aphthous stomatitis. Dermatol Clin. 14(2):243-56.

Reichart PA. (2000) Oral mucosal lesions in a representative cross-sectional study of aging Germans. Community Dent Oral Epidemiol. 28(5):390-8.

Revuz J, Guillaume JC, Janier M, Hans P, Marchand C, Souteyrand P, Bonnetblanc JM, Claudy A, Dallac S, Klene C. (1990) Crossover Study of Thalidomide vs Placebo in Severe Recurrent Aphthous Stomatitis. Arch Dermatol. 126(7):923-7.

Rivera-Hidalgo F, Shulman JD, Beach MM. (2004) The association of tobacco and other factors with recurrent aphthous stomatitis in an US adult population. Oral Dis. 10(6):335-45.

Rogers RS 3rd. (1997) Recurrent aphthous stomatitis: clinical characteristics and associated systemic disorders. Semin Cutan Med Surg. 16(4):278-83. Review.

Ruah CB, Stram JR, Chasin WD. (1988) Treatment of severe recurrent aphthous stomatitis with colchicine. Arch Otolaryngol Head Neck Surg. 114(6):671-5.

Saadoun D, Wechsler B, Terrada C, Hajage D, Le Thi Huong D, Resche- Rigon M, Cassoux N, Le Hoang P, Amoura Z, Bodaghi B, Cacoub P. (2010)

203

Azathioprine in severe uveitis of Behçet's disease. Arthritis Care Res (Hoboken). 62(12):1733-8.

Safadi RA. (2009) Prevalence of recurrent aphthous ulceration in Jordanian dental patients. BMC Oral Health. 22;9:31.

Sakuma S, Higashi Y, Sato N, Sasakawa T,Sengoku T, Ohkubo Y, Amaya T, Goto T. (2001) Tacrolimus suppressed the production of cytokines involved in atopic dermatitis by direct stimulation of human PMNC system (Comparison with steroids). International immunopharmacology J. 1(6):1219-26.

Sand FL, Thomsen SF. (2013) Efficacy and safety of TNF-α inhibitors in refractory primary complex aphthosis: a patient series and overview of the literature. J Dermatolog Treat. 24(6):444-6.

Savage NW, Mahanonda R, Seymour GJ, Bryson GJ, Collins RJ. (1988) The proportion of suppressor-inducer T lymphocytes is reduced in recurrent aphthous stomatitis. J Oral Pathol Med. 17(6):293-297.

Savage NW, Seymour GJ, Kruger BJ. (1986) I and Class II major histocompatibility complex antigens on epithelial cells in recurrent aphthous. Journal of Oral Pathology & Medicine. 15(4):191-5.

Sayinalp N, Ozcebe OI, Ozdemir O, Haznedaro ğlu IC, Dündar S, Kirazli S. (1996) Cytokines in Behçet's disease. J Rheumatol. 23(2):321-2.

Schreiner DT, Jorizzo JL. (1987) Behçet's disease and complex aphthosis. Dermatol Clin. 5(4):769-78.

Schroeder HE, Muller-Gluser W, Sallay K. (1984) Pathomorphologic features of the ulcerative stage of oral aphthus ulceration. Oral Surg Oral Med Oral Pathol. 58(3):293-305.

Scully C, Felix DH. (2005) Oral medicine, Update for the dental practitioner Aphthous and other common ulcers. Br Dent J. 10;199(5):259-264.

Scully C, Gorsky M, Lozada-NurF. (2003) The diagnosis and managements of recurrent aphthous ulcerations: a consensus approach.

204

J Am Dent Assoc. 134(2):200-207.

Scully C. (2006) Clinical practice. Aphthous ulceration. N Engl J Med. 13;355(2):165-72.

Scully C, Hodgson T, Lachmann H. (2008) Auto-inflammatory syndromes and oral health. Oral Dis. 14(8):690-9.

Sharma S, Ali FM, Saraf K, Mudhol A. (2014) Anti-helminthic drugs in recurrent apthous stomatitis: A short review. J Pharm Bioallied Sci. 6(2):65-68.

Sharquie KE, Najim RA, Al-Hayani RK, Al-Nuaimy AA, Maroof DM. (2008) The therapeutic and prophylactic role of oral zinc sulfate in management of recurrent aphthous stomatitis (ras) in comparison with dapsone. Saudi Med J. 29(5):734-8.

Sharquie KE. (1984) Suppression of Behçet's disease with dapsone. Br J Dermatol. 110(4):493-4.

Shemer A, Amichai B, Trau H, Nathansohn N, Mizrahi B, Domb AJ. (2008) Efficacy of a mucoadhesive patch compared with an oral solution for treatment of aphthous stomatitis. Drugs R D. 9(1):29-35.

Shigeaki Ohno, Masaki Ohguchi, Shigeto Hirose, Hidehiko Matsuda, Akemi Wakisaka, Miki Aizawa. (1982) Close Association of HLA-Bw51 With Behçet's Disease. Arch Ophthalmol. 100(9):1455-1458.

Ship II, Morris AW, Durocher RT, Burket WL. (1960) Recurrent aphthous ulcerations and recurrent herpes labialis in a professional school student population. Oral Surg Oral Med Oral Pathol. 13(10):1191-1202.

Ship II. (1965) Inheritance of aphthous ulcers of the mouth. J Dent Res. 44(5):837-844.

Ship II. (1972) Epidemiological aspect of recurrent aphthous ulcerations. Oral Surg Oral Med Oral Pathol. 33(3):400-406.

Shohat-Zabarski R1, Kalderon S, Klein T, Weinberger A. (1992) Close association of HLA-B51 in persons with recurrent aphthous stomatitis. Oral Surg Oral Med Oral Pathol. 74(4):455-8.

205

Sistig S, Cekic-Arambasin A, Rabatic S, Vucicevic-Boras V, Kleinheinz J, PiffkoJ. (2001) Natural immunity in recurrent aphthous ulceration. J of Oral Path Med. 30(5):275-80.

Sircus W. (1959) The Management of Recurrent Aphthous Stomatitis. Br Med J. 24;2(5155):804-6.

Sircus W, Church R, Kelleher J. (1957) Recurrent aphthous ulceration of the mouth. A study of the natural history, aetiology, and treatment. Quarterly Jour of Med. 26(102):235-249.

Sugi-Ikai N, Nakazawa M, Nakamura S, Ohno S, Minami M. (1998) Increased frequencies of interleukin-2- and interferon-gamma-producing T cells in patients with active Behcet's disease. Invest Ophthalmol Vis Sci. 39(6):996-1004.

Sullivan TP, King LE Jr, Boyd AS. (1998) Colchicine in dermatology. J Am Acad Dermatol. 39(6):993-9.

Sun A, Chiang CP, Chiou PS, Wang JT, Liu BY, Wu YC. Immunomodulation by levamisole in patients with recurrent aphthous ulcers or oral lichen planus. J Oral Pathol Med. 1994;23(4):172-7.

Sun A, Chang YF, Chia JS, Chiang CP. (2004) Serum interleukin-8 level is a more sensitive marker than serum interleukin-6 level in monitoring the disease activity of recurrent aphthous ulcerations. J Oral Pathol Med. 33(3):133-9.

Sun A, Wang JT, Chia JS, Chiang CP. (2006) Levamisole can modulate the serum tumor necrosis factor-alpha level in patients with recurrent aphthous ulcerations. J Oral Pathol Med. 35(2):111-6.

Sun A, Chia JS, Chang YF, Chiang CP. (2003) Levamisole and Chinese medicinal herbs can modulate the serum interleukin- 6 level in patients with recurrent aphthous ulcerations. J Oral Pathol Med. 32(4):206-14.

Sun A, Chia JS, Wang WB, Chiang CP. (2005) "Tien-Hsien liquid" can modulate antigen-stimulated cytokine production by T- cells isolated from patients with recurrent aphthous ulcerations. Am J Chin Med. 33(4):559-71.

206

Sun A, Wang YP, Chia JS, Liu BY, Chiang CP. (2009) Treatment with levamisole and colchicine can result in a significant reduction of IL-6, IL-8 or TNF-α level in patients with mucocutaneous type of Behcet’s disease. J Oral Path Med. 38(5):401–405.

Sun A, Chu CT, Liu BY, Wang JT, Leu JS, Chiang CP. (2000) Expression of interleukin-2 receptor by activated peripheral blood lymphocytes upregulated by plasma level of interleukin-2 in patients with recurrent aphthous ulcers. Proc Natl Sci Counc Repub China B. 24(3):116-22.

Sun Y, Schmitz JE, Acierno PM, Santra S, Subbramanian RA, Barouch DH, Gorgone DA, Lifton MA, Beaudry KR, Manson K, Philippon V, Xu L, Maecker HT, Mascola JR, Panicali D, Nabel GJ, Letvin NL. (2005) Dysfunction of simian immunodeficiency virus/simian human immunodeficiency virus-induced IL-2 expression by central memory CD4+ T lymphocytes. J Immunol. 15;174(8):4753-60.

Tappuni A.R, Kovacevic T, Shirlaw P, Challacombe SJ. (2005) Clinical Assessment of Disease Severity in Recurrent Aphthous Stomatitis. Oral Biosci Med. 2:5-13.

Tappuni A.R, Kovacevic T, Challacombe SJ. (2008) Is Topical or Systemic Treatment More Effective in Oral Ulceration. Oral Pathology & Oral Medicine: Pathogenesis of Disease. 2008 Queen Elizabeth II Conference (IADR). The Pan European Federation of the International Association for Dental Research.

Tappuni A.R, Kovacevic T, Shirlaw PJ, Challacombe SJ. (2013) Clinical assessment of disease severity in recurrent aphthous stomatitis. J Oral Path Med. 42(8):635-41.

Takahashi H, Suguro T, Okazima Y, Motegi M, Okada Y, Kakiuchi T. (1996) Inflammatory Cytokines in the Herniated Disc of the Lumbar Spine. Spine (Phila Pa 1976). 15;21(2):218-224.

Tasher D, Stein M, Dalal I, Somekh E. (2008) Colchicine prophylaxis for frequent periodic fever, aphthous stomatitis, pharyngitis and adenitis episodes. Acta Paeditr. 97(8):1090-2.

Taylor LJ, Bagg J, Walker DM, Peters TJ. (1992) Increased production of tumour necrosis factor by peripheral blood leukocytes in patients with recurrent oral aphthous ulceration. J Oral Pathol Med. 21(1):21-5.

Taylor LJ, Walker DM, Bagg J. (1993) A clinical trial of prostaglandin E2 in recurrent aphthous ulceration.

207

Br Dent J. 21;175(4):125-9.

Tezel A, Kara C, Balkaya V, Orbak R. (2009) An evaluation of different treatments for recurrent aphthous stomatitis and patient perceptions: Nd:YAG laser versus medication. Photomed Laser Surg. 27(1):101-6.

Thompson AC, Nolan A, Lamey PJ. (1989) Minor aphthous oral ulceration; a double-blind cross over study of beclomethasone dipropinate aerosal spray. Scottish Medical Jour. 34(5):531-2.

Thornhill MH, Baccaglini L, Theaker E, Pemberton MN. (2007) A randomized, double-blind, placebo-controlled trial of pentoxifylline for the treatment of recurrent aphthous stomatitis. Arch Dermatology. 143(4):463-70.

Umland SP, Nahrebne DK, Razac S, Beavis A, Pennline KJ, Egan RW, Billah MM. (1997) The inhibitory effects of topically active glucocorticoids on IL-4, IL-5, and interferon-gamma production by cultured primary CD4+ T cells. J Allergy Clin Immunol. 100(4):511-9.

Vallejo Garcia-Pola MJ, Martinez Diaz-Canel AI, Garcia Martin JM, Gonzalez Garcia M. (2002) Risk factors for oral soft tissue lesions in an adult Spanish population. Community Dent Oral Epidemio. 30(4):277-85.

Verity DH, Marr JE, Ohno S, Wallace GR, Stanford MR. (1999) Behçet's disease, the Silk Road and HLA-B51: historical and geographical perspectives. Tissue Antigens. 54(3):213-20.

Verpilleux MP, Bastuji-Garin S, Revuz J. (1999) Comparative analysis of severe aphthosis and Behçet's disease: 104 cases. Dermatology. 198(3):247-51.

Viguier M, Fouere S, Pascal F, Morel P. (2000) Herpetiform ulceration: 5 cases. Ann Dermatol Venereol. 127(8-9):707-10.

Volkov I, Rudoy I, Freud T, Sardal G, Naimer S, Peleg R, Press Y. (2009) Effectiveness of vitamin B12 in treating recurrent aphthous stomatitis: a randomized, double-blind, placebo-controlled trial. J Am Board Fam Med. 22(1):9-16.

Wallace GR, Pryce KM, Curnow SJ, Fortune F, Stewart JE, Modi N, Knight B, Stanford MR, Murray PI. (2005) Multiplex Bead Analysis of Cytokines in Serum From Patients With Behcet's Disease.

208

Invest Ophthalmol Vis Sci. 46 E-Abstract 5120.

Weckx LL, Hirata CH, Abreu MA, Fillizolla VC, Silva OM. (2009) Levamisole does not prevent lesions of recurrent aphthous stomatitis: a double-blind placebo-controlled clinical trial. Rev Assoc Med Bras. 55(2):132-8.

Woo MY, Cho O, Lee MJ, Kim K, Lee ES, Park S. (2012) Differential effects of colchicine in blood mononuclear cells of patients with Behçet disease in relation to colchicine responsiveness. Br J Dermatol. 167(4):914-21.

Wray D, Vlagopoulos TP, Siraganian RP. (1982) Food allergens and basophil histamine release in recurrent aphthous stomatitis. Oral Surg Oral Med Oral Pathol. 54(4):388-95.

Wray D. (1982) A double-blind trial of systemic zinc sulfate in recurrent aphthous stomatitis. Oral Surg Oral Med Oral Pathol. 53(5):469-72.

Xie W, Liu X, Xuan H, Luo S, Zhao X, Zhou Z, Xu j. (2006) Effect of betamethasone on neuropathic pain and cerebral expression of NF- kappa B and cytokines. Neuroscience Letters. 30;393(2-3):255-9.

Yalcindag FN, Yalcidag A, Batioglu F, Caglayan O. (2008) Evaluation of serum resistin levels in patients with ocular and non-ocular Behcet disease. Can J of Opthalmol. 43(4):473-5.

Yamamoto T, Yoneda K, Ueta E, Osaki T. (1994) Serum cytokines, interleukin-2 receptor, and soluble intercellular adhesion molecule-1 in oral disorders. Oral Surg Oral Med Oral Pathol. 78(6):727-35.

Yang TY, Jang TY. (2009) The value of local botulinum toxin A injection in the treatment of the pain of aphthous ulcer. Eur Arch Otorhinolaryngol. 266(3):445-8.

Yasui K, Kurata T, Yashiro M, Tsuge M, Ohtsuki S, Morishima T. (2010) The effect of ascorbate on minor recurrent aphthous stomatitis. Acta Paediatr. 99(3):442-5.

Yazici H, Pazarli H, Barnes CG, Tuzun Y, Ozyazgan Y, Silman A, Serdaroglu S, Oguz V, Yurdakul S, Lovatt GE, Yazici B, Somani S, Muftuoglu A. (1990) A controlled trial of azathioprine in Behcet's syndrome. N Engl Med. 1;322(5):281-5.

209

Yel L, Tezean I, Hasturk , Ersoy F, Sanal O, Yavazyilmaz E. (1994) Oral findings, treatment and follow up of a case with major aphthous stomatitis (Sutton's disease). J Clin Pediat Dent. 19(1):49-53.

Yeoman CM, Greenspan JS, Harding SM. (1978) Recurrent oral ulceration. A double-blind comparison of treatment with betamethasone valerate aerosol and placebo. Br Dent J. 21;144(4):114-6.

Ylikontiola L, Sorsa T, Häyrinen-Immonen R, Salo T. (1997) Doxymycine-cyanoacrylate treatment of recurrent aphthous ulcers. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 83(3):329-33.

Zadik Y, Levin L, Shmuly T, Sandler V, Tarrasch R. (2012) Recurrent aphthous stomatitis: stress, trait anger and anxiety of patients. J Calif Dent Assoc. 40(11):879-83.

Zain RB. (2000) Oral recurrent aphthous ulcers/stomatitis: prevalence in Malaysia and an epidemiological update. J Oral Sci. 42(1):15-9.

Zand N, Fateh M, Ataie-Fashtami L, Djavid GE, Fatemi SM, Shirkavand A. (2012) Promoting wound healing in minor recurrent aphthous stomatitis by non- thermal, non-ablative CO(2) laser therapy: a pilot study. Photomed Laser Surg. 30(12):719-23.

Zierhut M, Mizuki N, Ohno S, Inoko H, Gul A, Onoe K. (2003) Immunology and functional genomics of Behcet’s disease. Cell Mol Life Sci. 60(9):1903-22.

Zemer D, Livneh A, Danon YL, Pras M, Sohar E. (1991) Long term colchicine treatment in children with familial Mediterranean fever. Arthritis Rheum. 34(8):973-7.

Zhou Y, Chen Q, Meng W, Jiang L, Wang Z, Liu J, Lin M, Zhou H, Chen X, Zhao M, Zeng X. (2010) Evaluation of penicillin G potassium troches in the treatment of minor recurrent aphthous ulceration in a Chinese cohort: a randomized, double- blinded, placebo and no-treatment-controlled, multicenter clinical trial. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 109(4):561-6.

Zöller NN, Kippenberger S, Thaçi D, Mewes K, Spiegel M, Sättler A, Schultz M, Bereiter-Hahn J, Kaufmannand R, Bernd A. (2008) Evaluation of beneficial and adverse effects of glucocorticoids on a newly developed full-thickness skin model. Toxicology In Vitro. 22(3):747-59.

210

211

212