Effectiveness of bronchial thermoplasty in patients with severe refractory asthma: Clinical and histopathologic correlations

Accepted Manuscript Effectiveness of bronchial thermoplasty in patients with severe refractory asthma: clinical and histopathological correlations Marina Pretolani, Pharm. D., Ph.D., Anders Bergqvist, Ph.D, Gabriel Thabut, M.D., Ph.D., Marie-Christine Dombret, M.D., Dominique Knapp, Fatima Hamidi, MS, Loubna Alavoine, M.D., Camille Taillé, M.D., Ph.D., Pascal Chanez, M.D., Ph.D., Jonas S. Erjefält, Ph.D., Michel Aubier, M.D. PII:

S0091-6749(16)30896-X

DOI:

10.1016/j.jaci.2016.08.009

Reference:

YMAI 12322

To appear in:

Journal of Allergy and Clinical Immunology

Received Date: 24 April 2016 Revised Date:

8 July 2016

Accepted Date: 8 August 2016

Please cite this article as: Pretolani M, Bergqvist A, Thabut G, Dombret M-C, Knapp D, Hamidi F, Alavoine L, Taillé C, Chanez P, Erjefält JS, Aubier M, Effectiveness of bronchial thermoplasty in patients with severe refractory asthma: clinical and histopathological correlations, Journal of Allergy and Clinical Immunology (2016), doi: 10.1016/j.jaci.2016.08.009. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Effectiveness of bronchial thermoplasty in patients with severe

2

refractory asthma: clinical and histopathological correlations

3 Marina Pretolani, Pharm. D., Ph.D.

5

Gabriel Thabut, M.D., Ph.D.

6

Dominique Knapp

a,b,c

d

, Anders Bergqvist, Ph.D ,

a,b,c,d,e,g

, Marie-Christine Dombret, M.D.

a,b,c,g

, Fatima Hamidi, MS

a,b,c

a,b,c,f,g

,

h

RI PT

4

, Loubna Alavoine, M.D. ,

7

Camille Taillé, M.D., Ph.D. a,b,c,f,g, Pascal Chanez, M.D., Ph.D. i, Jonas S Erjefält, Ph.D. d,

8

and Michel Aubier, M.D. a,b,c,f,g a

10

Inserm UMR1152, Physiopathology and Epidemiology of Respiratory Diseases, Paris,

SC

9

France; b

Paris Diderot University, Faculty of Medicine, Bichat campus, Paris, France;

12

c

Laboratory of Excellence, INFLAMEX, Université Sorbonne Paris Cité and DHU FIRE,

13

M AN U

11

Paris, France;

14

d

Unit of Airway Inflammation, Lund University, Lund, Sweden;

15

e

Departments of Pneumology B and f Pneumology A, Bichat-Claude Bernard University

16

Hospital, Paris, France; g

Assistance Publique des Hôpitaux de Paris, Paris, France;

18

h

Clinical Investigation Center, Bichat-Claude Bernard University Hospital, Paris, France;

19

i

Inserm U1067 and CNRS UMR7733, Department of Respiratory Diseases, APHM Aix-

20

TE D

17

Marseille University, Marseille, France.

21

M. Pretolani and A. Bergqvist contributed equally to this work.

23

Corresponding author: Prof. Michel Aubier, M.D., Service de Pneumologie A, Hôpital

24

Bichat-Claude

25

[email protected]; phone: (+33) 1 40 25 68 00; fax: (+33) 1 40 25 88 08.

26

Sources of support: This study was supported, in part, by Boston Scientific, Marlborough,

27

MA, USA

28

Running head: Bronchial thermoplasty and severe asthma

29

Word count in the body of the manuscript (excluding abstract and references):

30

This article has an Online Repository at www.jacionline.org

EP

22

AC C

Bernard,

46

rue

Henri

Huchard,

75018

Paris,

France;

e-mail:

2

ACCEPTED MANUSCRIPT Abstract

32

Background. The effectiveness of bronchial thermoplasty (BT) has been reported in severe

33

asthma, yet its impact on the different bronchial structures remains unknown.

34

Objective. To examine the effect of BT on bronchial structures and to explore their

35

association with clinical outcome in severe refractory asthmatics.

36

Methods. Bronchial biopsies (n = 300) were collected from 15 severe uncontrolled

37

asthmatics before and 3 months after BT. Immunostained sections were assessed for airway

38

smooth muscle (ASM) area, sub-epithelial basement membrane thickness, nerve fibers and

39

epithelium neuroendocrine cells. Histopathological findings were correlated with clinical

40

parameters.

41

Results. BT significantly improved asthma control and quality of life at both 3 and 12 months

42

and decreased the numbers of severe exacerbations and the dose of oral corticosteroids. At

43

3 months, this clinical benefit was accompanied by a reduction in ASM area (median values

44

[25-75 IQR] before and after BT, respectively, 19.7% [15.9-22.4] and 5.3% [3.5-10.1], P <

45

0.001), in sub-epithelial basement membrane thickening (4.4 µm [4.0-4.7] and 3.9 µm [3.7-

46

4.6], P = 0.02), in sub-mucosal nerves (1.0 ‰ immunoreactivity [0.7-1.3] and 0.3 ‰

47

immunoreactivity [0.1-0.5], P < 0.001), in ASM-associated nerves (452.6 immunoreactive

48

pixels per mm2 [196.0-811.2] and 62.7 immunoreactive pixels per mm2 [0.0-230.3], P = 0.02)

49

and in epithelium neuroendocrine cells (4.9 per mm2 [0-16.4] and 0.0 per mm2 [0-0], P =

50

0.02). Histopathological parameters were associated with asthma control test, number of

51

exacerbations, and visits to emergency department (all P ≤ 0.02), 3 and 12 months after BT.

52

Conclusion. BT is a treatment option in severe therapy-refractory asthma that down-

53

regulates selectively structural abnormalities involved in airway narrowing and bronchial

54

reactivity, particularly ASM, neuroendocrine epithelial cells and bronchial nerve endings.

55

Word count in the abstract: 274

56

Key words: Refractory asthma; asthma control; airway smooth muscle; airway remodeling;

57

epithelium neuroendocrine cells; mucosal nerves; bronchial epithelium

AC C

EP

TE D

M AN U

SC

RI PT

31

3

ACCEPTED MANUSCRIPT CAPSULE SUMMARY

59

Bronchial thermoplasty improved disease control in highly symptomatic therapy-uncontrolled

60

severe asthmatics and this clinical efficacy was associated with a reduction in airway smooth

61

muscle area, in neuroendocrine epithelial cells and in sub-mucosal nerves.

62

CLINICAL IMPLICATION STATEMENT

63

Bronchial thermoplasty is an effective treatment for severe refractory therapy-uncontrolled

64

asthma and its clinical benefit is related to alterations in specific airway structures induced by

65

this procedure.

66

Abbreviations

67

IL, interleukin; BT, bronchial thermoplasty; ASM, airway smooth muscle; FEV1, Forced

68

Expiratory Volume in one second; FVC, Forced Vital Capacity; LABA, long-lasting β2-

69

agonists; ICS, inhaled corticosteroids, OCS, oral corticosteroids, ACT, Asthma Control Test;

70

AQLQ, Asthma Quality Control Questionnaire; ICU, Intensive Care Unit; SBM, sub-epithelial

71

basement membrane; IQR, interquartile range.

AC C

EP

TE D

M AN U

SC

RI PT

58

4

ACCEPTED MANUSCRIPT Introduction

73

Severe asthma is characterized by persistent symptoms and frequent exacerbations that

74

contribute to the mortality and morbidity associated with this disease (1). The heterogeneity

75

of airway inflammation in severe asthma has led to the recognition of multiple distinct

76

endotypes allowing the guidance of use of novel specific biologics, such as anti-IgE and anti-

77

interleukin (IL)-5 monoclonal antibodies (2,3). These biologics are indicated for patients

78

whose asthma remains uncontrolled, despite high doses of inhaled corticosteroids (ICS),

79

long-acting bronchodilators (LABA) and oral corticosteroids (OCS) (4,5). However, some

80

patients experience severe exacerbations despite treatment with these biologics, or even

81

with undetectable inflammation (6).

82

Recent studies conducted in asthmatic children demonstrated that structural changes of the

83

airway wall, known under the term of ‘airway remodeling’, arise independently on

84

inflammation (7). Remodeling in asthma is manifested as epithelial cell hyperplasia, goblet

85

cell metaplasia, sub-epithelial fibrosis, angiogenesis and increase in airway smooth muscle

86

(ASM) mass (8,9). These changes correlate with asthma severity, and with the degree of

87

airflow obstruction (8,9).

88

Bronchial thermoplasty (BT) is an endoscopic procedure that targets primarily airway

89

remodeling by delivering temperature-controlled radio frequency energy to the airway wall.

90

Clinical trials of BT in patients with moderate to severe asthma have reported improvement

91

of quality of life and reduction in the number of exacerbations (10-12). However, which

92

patients could benefit from BT is still unknown and the mechanisms underlying clinical

93

improvement are currently under debate (13,14). A reduction of ASM mass following BT has

94

recently been demonstrated in patients with severe uncontrolled asthma (15), yet the

95

relationship with clinical benefit remains elusive. Furthermore, since heat energy produced

96

during BT can potentially alter airway structural components other than the ASM, additional

97

mechanisms may contribute to the observed clinical efficacy.

98

The current study addresses these important issues, by performing a detailed quantitative

99

analysis of major structural airway alterations and of mucosal inflammation in severe

AC C

EP

TE D

M AN U

SC

RI PT

72

5

ACCEPTED MANUSCRIPT

asthmatics undergoing BT and by correlating the observed modifications with key clinical

101

outcomes.

102

Methods

103

Subjects. Fifteen adult patients (27-69 years of age) who met the American Thoracic Society

104

criteria for severe refractory asthma ATS/ERS criteria (16) were recruited at the Pneumology

105

A department of the Bichat University Hospital (Paris, France) between January and

106

December 2013 (Table 1). Selection criteria were: subjects with uncontrolled severe asthma,

107

assessed by score on asthma control test (ACT) ≤ 15, despite optimal management and

108

maximal medications for at least 12 months before entry, pre-bronchodilator forced

109

expiratory volume in one second (FEV1) > 30% and < 80% of predicted; at least 3

110

exacerbations, defined as a worsening of asthma symptoms requiring treatment with OCS,

111

during the previous year. Ten patients met the criteria for receiving the anti-IgE monoclonal

112

antibody, omalizumab, for 6 months before entry in the protocol without successful clinical

113

outcome. For these 10 patients discontinuation of omalizumab occurred at least 6 months

114

before the first BT session (patient’s details are provided in this article’s Online Repository at

115

www.jacionline.org).

116

Before, 3 and 12 months after BT, all subjects underwent assessment of FEV1 and forced

117

vital capacity (FVC), before and after the inhalation of 400 µg of salbutamol. Scores on ACT

118

(response scale from 1 to 25) and of asthma quality of life questionnaire (AQLQ) (response

119

scale from 1 to 7) (17-19), number of exacerbations, hospitalizations for asthma and in

120

intensive care unit (ICU), visits to emergency department, as well as treatments were

121

recorded throughout the study.

122

The protocol was approved by the CPP Ile-de-France I Ethics Committee (n° 2012-sept-

123

13003) and all subjects gave their written informed consent. This trial is registered on

124

ClinicalTrials.gov NCT01777360.

125

BT procedure. BT was performed using the ALAIR® System (Boston Scientific,

126

Marlborough, MA), as previously described (10-12) (details are provided in the Online

AC C

EP

TE D

M AN U

SC

RI PT

100

6

ACCEPTED MANUSCRIPT

Repository). All patients underwent 3 sessions of BT, separated by one-month intervals (15).

128

Median numbers (25-75 interquartile range, IQR) of 44 (37-57), 57 (52-64) and 46 (43-78)

129

heat-activations were administered during the first, second and third BT session, in the right

130

lower lobe, the left lower lobe and the two upper lobes respectively. No heat-activation was

131

delivered in the middle lobe (15).

132

Bronchoscopy and biopsy collection. A bronchoscopy was performed 15 days before the

133

first BT procedure, and 3 months after the last one by the same operator (M-C.D.) (15,20).

134

Ten bronchial biopsies were collected at the same locations, i.e., 3 biopsies from the right

135

lower lobe (B7-B8, B8-B9 and B9-B10), 2 biopsies from each upper lobe (B1-B2 and lingula),

136

2 biopsies in the middle lobe (B4-B5) and 3 biopsies from the left lower lobe (B7-B8, B8-B9,

137

B9-B10) (15).

138

Histopathological studies. Histopathological analyses were performed blinded at the Unit

139

of Airway Inflammation, Lund University, Sweden (methodological details are provided in the

140

Online Repository). Briefly, automated immunohistochemistry was performed, as previously

141

described (21,22), using a Dako Cytomation staining robot and the antibodies listed in Table

142

E1. Stained sections were digitalized and analyzed with computerized image analysis

143

software. ASM area was manually delineated under guidance of α-smooth-muscle actin

144

immunostaining (15,20). Neuroendocrine epithelial cells, defined by their distinct PGP9.5-

145

positive nuclei, were manually counted and normalized to the epithelial surface area.

146

Immunoreactivity for PGP9.5-positive nerve fibers, blood and lymph vessels, eosinophils and

147

neutrophils was calculated by dividing the positive staining for each marker (as determined

148

by a positive pixel count algorithm) with the total tissue area (total pixels). Epithelial

149

morphological status was qualitatively assessed on haematoxylin stained sections, as

150

described previously (23). Finally, collagen was stained with Masson’s trichrome dye and its

151

extent and intensity were quantified. Sub-epithelial basement membrane (SBM) thickening

152

was determined by multiple point-to-point measurements, with approximately 50-µm intervals

153

between each measurement.

AC C

EP

TE D

M AN U

SC

RI PT

127

7

ACCEPTED MANUSCRIPT

All examinations, which were performed on blinded sections, were assessed in the 10 lung

155

biopsies obtained before BT and values were compared to those obtained in the 8 biopsies

156

from the BT-treated lung lobes at 3 months post-treatment and to the 2 biopsies from the BT-

157

untreated middle lobe. Decoding of tissue sections took place once all analyses were

158

completed.

159

In separate analyses, values obtained at 3 months in biopsies from the middle BT-untreated

160

lung lobe were compared to those obtained at the same time in the BT-treated lung lobes

161

and to values detected in the 10 biopsies collected before BT.

162

Statistical analyses Qualitative variables are shown as numbers and percentages.

163

Quantitative variables are reported as mean values and standard deviations, except where

164

otherwise indicated. We used linear regression models to compare the values of biological

165

and clinical variables measured at the 3 time points: before, 3 months and 12 months after

166

BT. When the normality and the homogeneity of variance assumptions were not satisfied,

167

variables were replaced by their log-transformed values. The number of events was

168

compared between study groups with the use of Poisson regression with correction for

169

treatment exposure and over dispersion. The models described above included both fixed

170

(time) and random (patient) effects to account for serial measurements made in the same

171

patients (mixed effect models). Since some variables included a large proportion of zero

172

values, pattern mixtures models were used. In addition, because multiple hypotheses were

173

tested, the Benjamini and Hochberg procedure has been used to adjust P values for multiple

174

comparisons. Potential associations between histochemical and morphometric variables

175

were assessed using the nonparametric Spearman’s rank correlation (r). All tests were two-

176

tailed; P values less than 0.05 were considered significant. Data were analyzed statistically

177

using R 2.15.2 (R Foundation for Statistical Computing, Vienna, Austria) software.

AC C

EP

TE D

M AN U

SC

RI PT

154

8

ACCEPTED MANUSCRIPT Results

179

Clinical effects of BT

180

Clinical and airway functional parameters were examined before and 3 and 12 months after

181

BT (Table 1). Although 6 out of the 15 BT-treated severe asthmatics still experienced

182

uncontrolled asthma at 3 months, we found overall significantly higher scores on ACT and

183

AQLQ (P < 0.001 for both comparisons), lower number of severe exacerbations (P < 0.001),

184

of visits to emergency department (P < 0.001), of hospitalization for asthma (P < 0.01), and

185

in ICU (P < 0.001), than those measured before BT (Table 1).

186

The clinical benefit observed at 3 months persisted at 12 months (Table 1). Indeed, a

187

significant decrease in the number of exacerbations (P < 0.001), of hospitalization for asthma

188

(P < 0.001) and in ICU (P = 0.008), of visits to emergency department (P < 0.001), and a rise

189

in ACT and AQLQ scores (P < 0.001 for both comparisons), were observed at 12 months, as

190

compared to values obtained before BT. In addition, at 12 months, 8 out of the 10 patients

191

still required regular OCS, and the mean daily dose of oral prednisone was significantly lower

192

as compared to that administered at entry (13.8 versus 31.5 mg per day 12 months after, as

193

compared to before BT, P = 0.002, Table 1).

194

Twelve months after BT, 4 out of the 15 patients still had uncontrolled asthma, with a mean

195

ACT score of 7.8 and an AQLQ score of 2.4 (Table E2). Patients were considered BT-

196

unresponsive when ACT at 3 and 12 months after BT remained lower than 15. Although in

197

these 4 poorly responsive patients BT decreased the number of severe exacerbations by

198

approximately 80%, values remained statistically higher than those observed in BT-

199

responsive asthmatics (P < 0.01, Table E2). The 4 poor responsive patients still required

200

OCS 12 months after BT at a daily dose significantly higher than the 6 patients among the 11

201

responders that were still on OCS (Table E2).

202

BT was not associated with changes in blood eosinophils and pulmonary function tests at

203

both 3 and 12 months (Table 1).

AC C

EP

TE D

M AN U

SC

RI PT

178

9

ACCEPTED MANUSCRIPT Histopathological effects of BT

205

At baseline, ASM area ranged from 9.1 to 30.3 % of the mucosal area (median [25-75 IQR] =

206

19.7 % [16.2-21.8]). BT resulted in a significant decrease of ASM area at 3 months, with a

207

median [25-75 IQR] value of 5.2 % [3.7-9.8] (P < 0.001) (Fig 1A, exemplified in Fig 2A-B).

208

This decrease was accompanied by a significant larger biopsy area stained for collagen and

209

by an intensification of its expression (Fig E4).

210

We then determined whether BT altered SBM thickening, blood and lymphatic vessels, sub-

211

mucosal and ASM-associated nerve fibers, and mucosal granulocytes (Fig 1A to 1I).

212

BT marginally, but significantly, decreased SBM thickening (median values before and 3

213

months after BT of 4.4 and 3.9, respectively, P = 0.02, Fig 1B), without modifying significantly

214

the density of blood and lymphatic vessels (Fig 1C and 1D). Sub-mucosal and ASM-

215

associated nerve fibers were significantly reduced 3 months after BT, as compared to values

216

measured before BT. Indeed, sub-mucosal nerves amounted to 1.0 ‰ immunoreactivity [0.7-

217

1.3] and to 0.3 ‰ immunoreactivity [0.1-0.5], before and after BT, respectively, P < 0.001,

218

and values of of ASM-associated nerves were of 452.6 per mm2 [196.0-811.2] and of 62.7

219

per mm2 [0.0-230.3], before and after BT, respectively P = 0.02) (Fig 1E and 1F).

220

Sub-epithelial mucous glands (Fig 1G), eosinophils and neutrophils (Fig 1H and 1I) were not

221

significantly modified by BT.

222

Finally, we examined whether BT targeted the bronchial epithelium (Fig 1J to 1O). Neither

223

the proportion of regenerating bronchial epithelium, normal stratified columnar, metaplastic,

224

or squamous epithelium (Fig 1J, 1K, 1L and 1M), nor goblet cell hypertrophy/hyperplasia (Fig

225

1N) were modified by BT. In contrast, we observed a significant reduction (P = 0.02) in the

226

numbers of epithelial neuroendocrine PGP+ cells (Fig 1O). Hence, whereas these cells were

227

clearly present in the bronchial epithelium from 11 out of the 15 severe asthmatics at the

228

inclusion, with median value (25-75 IQR) of 4.9 (0.3-14.1) cells per mm2 of epithelium layer,

229

they were detectable only in 2 patients, 3 months after BT (Fig 1O).

230

We previously reported that BT decreased ASM area also in the heat-untreated middle lobe

231

in 50 to 70% of severe asthmatics (15). Therefore, we determined whether this same effect

AC C

EP

TE D

M AN U

SC

RI PT

204

10

ACCEPTED MANUSCRIPT

was observed on the hallmarks of airway inflammation and remodeling currently investigated

233

(Table E3). We found a trend towards a decrease in sub-mucosal and ASM-associated

234

nerves, and in the proportion of glandular tissue in the BT-untreated middle lobes (Table E3).

235

A significant reduction in the number of neuroendocrine epithelial cells (P = 0.04) was also

236

observed in the BT-untreated middle lobes at 3 months, as compared to values obtained in

237

all lung lobes before BT (Table E3). In these analyses, 4 variables were characterized by a

238

large amount of zero values in addition to continuous positive values: glandular tissue,

239

number of PGP+ neuroendocrine cells, regenerating epithelium and squamous metaplastic

240

epithelium. The use of pattern mixture models, using logistic and truncated Poisson

241

distributions instead of non parametric tests yielded similar results for these 4 variables

242

(Table E3).

243

Because 4 of the 15 severe asthmatics showed limited clinical improvement upon BT at 12

244

months, we compared the degree of their airway remodeling at 3 months to that observed in

245

BT-responsive patients (Fig 3). BT-poorly-responsive asthmatics had a median value of ASM

246

area of 14.6 %, as compared to 5.7 % observed in the remaining 11 severe asthmatics (P =

247

0.05, Fig 3A). Sub-mucosal and ASM-associated nerves were similarly down-regulated in

248

BT-responsive and poorly responsive patients (Fig 3C and 3D) and no detectable differences

249

were found in SBM thickening (Fig 3B) and in other hallmarks of airway remodeling, or in

250

granulocyte infiltration (data not shown). Finally, median numbers of neuroendocrine cells

251

per mm2 of bronchial epithelium was 3.2 in the 4 BT-poorly responsive patients, whereas

252

these cells were undetectable in the rest of the patients (P = 0.05).

253

Correlation analyses

254

Multiple correlation analyses conducted before and 3 months after BT demonstrated

255

significant positive associations between ASM area and sub-mucosal and ASM-associated

256

nerves and epithelium neuroendocrine cells (Table E4). The highest significant correlation

257

was observed by comparing the extent of ASM area and the number of submucosal nerves

258

and of epithelium neuroendocrine cells (r=0.56; 95%CI: 0.15-0.97 and r=0.75; 95%CI: 0.2-1,

AC C

EP

TE D

M AN U

SC

RI PT

232

11

ACCEPTED MANUSCRIPT respectively).

260

Changes in histopathological parameters induced by BT were not associated with the

261

cumulative numbers of heat-stimulations administered during the 3 sessions (Table E5).

262

We then investigated the relationship between the components of airway remodeling that

263

were altered 3 months after BT and clinical parameters at 3 and 12 months (Table 2 and

264

Tables E6 to E10). Three months after BT, ASM highly significantly correlated with ACT

265

score (P = 0.003) and with the number of severe exacerbations (P < 0.001), of visits to

266

emergency department (P = 0.003) and of hospitalization for asthma (P = 0.03) (Table 2).

267

Similar associations were seen when considering sub-mucosal and ASM-associated nerves,

268

and number of epithelium neuroendocrine cells, whereas SBM thickening correlated

269

exclusively with ACT score (Table 2). In a prospective analysis, we found that most of these

270

correlations were maintained at 12 months (Table 2).

271

The histopathologic parameters presently investigated failed to correlate with pre- and post-

272

bronchodilator FEV1, FVC, FEV1/FVC at 3 months (Tables E6 to E10).

273

Discussion

274

The present study was aimed at investigating the clinical benefit of BT in patients with severe

275

refractory asthma and at determining which alterations in the different bronchial structures

276

were associated with clinical improvement.

277

BT improved significantly asthma control, assessed by daily symptoms (ACT) (+ 52 %), rate

278

of severe exacerbations, hospitalization for asthma, ICU stays, and visits to emergency

279

room. These effects were accompanied by a significant reduction in the maintenance doses

280

and in the number of bursts of OCS and improved in AQLQ (+ 62 %) scores at 12 months.

281

Clinical benefit was detectable at 3 months after BT and persisted until 12 months. Although

282

supporting, in part, earlier observations found in patients with severe asthma (10-12, 24,25),

283

it should be emphasized that our severe asthmatics were more symptomatic than those

284

enrolled in earlier studies, with mean rates of exacerbations during the year before entry of

285

9.7, instead of 0.7, mean AQLQ score of 2.6, instead of 4.7 and greater prevalence of

AC C

EP

TE D

M AN U

SC

RI PT

259

12

ACCEPTED MANUSCRIPT

maintenance use of OCS (67 %, instead of 41%) (11,12). Therefore, the clinical improvement

287

at 12 months currently reported in the present study was of a higher magnitude than that

288

previously shown, with a 92 % decrease in the number of exacerbations, 90 % lower number

289

of visits to emergency room, 88 % less hospitalizations for asthma and in ICU, and 93 % and

290

62 % improvement in ACT and AQLQ scores, respectively. Since we did not include a sham

291

control group in the current study, we cannot rule out that subjects became more adherent

292

to treatments upon BT, as compared to the 12 months before enrollment. However, this is

293

very unlikely because all the patients were part of the French longitudinal COBRA cohort, in

294

which a follow-up is scheduled every 6 months. As shown in Table E11, the clinical

295

characteristics (ACT, exacerbations…) of 20 severe asthmatics belonging to the COBRA

296

cohort are identical to the 15 subjects who underwent the 3 BT sessions. In these 20 severe

297

asthmatics who could be considered as a control group, clinical parameters remained stable

298

over one year, making unlikely an effect of drug adherence to explain the clinical

299

improvement observed after BT.

300

Histopathological examination of bronchial biopsies collected 3 months after BT showed a

301

significant reduction of the ASM area (73 %), thereby extending our previous results (15) and

302

confirming those recently obtained by Chakir et al. (26). This reduction in ASM area

303

correlated statistically with the improvement of asthma control, quality of life, and the

304

decrease of severe exacerbations, hospitalization visits to emergency department for

305

asthma, at 3 months. Importantly, all correlations were maintained at one year after BT. No

306

association between clinical improvement and histological parameters was reported by

307

Chakir et al (26). In this report, bronchial biopsies were collected in 17 subjects during the

308

first BT session in the non-treated lobe (the left lower lobe), BT being performed in the right

309

lower lobe. During their second treatment, which occurred 3-14 weeks (median, 3 wk), all 17

310

subjects underwent biopsies of the previously treated right lobe. However, only 9 out of 17

311

patients underwent biopsies of the right lower lobe during the third BT procedure (7-22

312

weeks, median 8 wk). In our current study, all subjects underwent systematic biopsies (10 in

313

all lung lobes) before and 3 months after the third BT procedure. This may explain why

AC C

EP

TE D

M AN U

SC

RI PT

286

13

ACCEPTED MANUSCRIPT

Chakir et al failed to find correlations between BT-induced changes in ASM area and clinical

315

improvement at one year. We also found that collagen deposition in the bronchial sub-

316

mucosa (both the extent and the intensity of the expression) increased in parallel with the

317

reduction of ASM area. Whether this extracellular matrix synthesis occurs in response to BT-

318

induced tissue injury and its precise composition remain to be determined. Of note, Chakir et

319

al. (26) reported a decrease in type 1 collagen deposition underneath SBM after BT. Further

320

studies are needed to determine which types of collagen are affected by BT and its precise

321

localization in the bronchial sub-mucosa.

322

Supporting previous observations (11,12,24), BT failed to alter pre- and post-bronchodilator

323

FEV1 and FVC.

324

Studies conducted in dogs and in patients scheduled for lung resection have shown that BT

325

selectively ablated ASM (27,28). In the current report, we found that BT slightly, but

326

significantly reduced SBM thickening, without altering blood and lymphatic vessels and

327

mucous glands, indicating that this procedure did not target other key components of airway

328

remodeling associated with severe asthma (20, 29-30).

329

The parasympathetic nervous system is an important constrictor neural pathway that controls

330

airway tone (31). Stimulation of cholinergic nerves causes bronchoconstriction, mucus

331

secretion, and bronchial vasodilation (32). We found that autonomic nerve fibers, both in the

332

bronchial sub-mucosa and within the ASM bundles, were drastically decreased 3 months

333

after BT. This reduction was significantly associated with the decline in the number of severe

334

exacerbations, suggesting that damage of autonomic-innervated structures induced by BT

335

down-regulated airway excitability and thus improved asthma control.

336

Because the bronchial epithelium could be damaged by the heat shock induced by BT, we

337

performed a morphological analysis of the epithelium as previously done (23) before and 3

338

months after BT. Neither epithelial regeneration, epithelial metaplasia, nor the extent of

339

normal columnar stratified epithelium, was modified by BT. These findings indicate that, if

340

epithelial injury occurred, it was very early after heat-delivery and that the absence of visible

AC C

EP

TE D

M AN U

SC

RI PT

314

14

ACCEPTED MANUSCRIPT

morphological alterations observed at 3 months suggests that the repair process has already

342

been completed. Although BT may alter functional capacities of the bronchial epithelium

343

without modifying visually its structure, this hypothesis remains to be explored.

344

Neuroendocrine cells are specialized epithelial cells distributed throughout the bronchial tree

345

as solitary cells, or as innervated clusters (neuro-epithelial bodies) (33). These cells drive

346

fetal lung growth and differentiation through the paracrine secretion of bioactive amines,

347

neuropeptides and growth factors, including serotonin, substance P, neurokinin A, bombesin

348

and calcitonin gene-related peptide (33). In addition, they function as intrapulmonary

349

hypoxia-sensitive chemoreceptors, thereby contributing to the control of airway tone (33,34).

350

Although the presence of pulmonary neuroendocrine cells in normal airways has been

351

illustrated (34), their possible role in the pathogenesis of asthma has not been clearly

352

demonstrated. We found that BT decreased by approximately 95% the number of

353

neuroendocrine epithelial cells at 3 months. This decrease highly correlated with the

354

improvement of asthma control (ACT and AQLQ scores), number of exacerbations,

355

hospitalization for asthma and in ICU and visits to emergency room. These data indicate that

356

the epithelium neuroendocrine cells are also a valuable hallmark of the beneficial clinical

357

effects of BT.

358

Surprisingly, we also found that the number of neuroendocrine cells was down-regulated in

359

the middle lobe, which was not treated by BT. However, as reported in a previous study (15),

360

no significant decrease in ASM area in the non-treated middle was noted. It should be noted,

361

however, that we currently expressed ASM area as percentage of the sub-mucosal tissue

362

area, whereas in our previous report (15), the surface area of ASM was calculated as a

363

percentage of the total biopsy area. This, and the fact that more patients have been included

364

in this study, may explain some variation between our two studies. Furthermore, in the

365

middle lobe, only two biopsies were taken before and after BT, whereas for the treated lobes,

366

8 biopsies were analyzed before and after BT. Further studies with higher numbers of

AC C

EP

TE D

M AN U

SC

RI PT

341

15

ACCEPTED MANUSCRIPT

patients and similar morphological approaches are needed to determine the effect of BT on

368

the non-treated middle lung lobe.

369

In a sub-analysis, we observed that the favorable clinical outcome following BT was not

370

consistent in the 15 patients. Indeed, after 12 months, 4 out of the 15 patients continued to

371

experience poor asthma control, as attested by values of ACT and AQLQ scores similar to

372

those measured at entry. This finding may be explained by distinct intrinsic sensitivities of

373

severe asthmatics to BT, despite apparent clinical similarities, or by an incomplete

374

effectiveness of the procedure on the different airway structures. Indeed, our results showing

375

that lower asthma control in response to BT was accompanied by an impaired reduction in

376

ASM area and in the number of neuroendocrine epithelial cells, suggest a causal relationship

377

between the ablation of these bronchial structural elements and clinical benefit.

378

Interestingly, numbers of mucosal eosinophils and neutrophils were unchanged, suggesting

379

that clinical improvement induced by BT is unrelated to an effect on granulocytic

380

inflammation. In a recent study, Darcy et al (36) reported a decrease in the bronchoalveolar

381

lavage fluid of transforming-growth factor β1 and of CCL5, 3 and 6 weeks after BT, whereas

382

IL-4, IL-5 and IL-17 were unaffected. However, only 11 patients were included in this study,

383

which makes difficult the correlation between these changes and clinical improvement

384

observed after BT to be assessed.

385

In conclusion, this is the first study that provides a link between the extent of ASM reduction

386

induced by BT and improvement of disease control in highly symptomatic therapy-

387

uncontrolled severe asthmatics. Importantly, apart from ASM, we identified neuroendocrine

388

epithelial cells, sub-mucosal nerves and, to a lesser extent, SBM thickening, as important

389

targets whose reduction after BT correlated with the clinical efficacy.

390

Whether these airway structural abnormalities need to be down-regulated simultaneously by

391

BT to deliver a therapeutic advantage, remains to be answered. At this stage, our results

392

support BT as an effective treatment option for severe refractory therapy-uncontrolled

AC C

EP

TE D

M AN U

SC

RI PT

367

16

ACCEPTED MANUSCRIPT

asthma and suggest that the clinical benefit is related to alterations in specific airway

394

structures induced by the procedure.

395

Acknowledgments

396

We thank the patients who made this study possible, Olivier Thibaudeau (Morphologic

397

platform, Inserm UMR1152) for biopsy inclusion and H&E staining and the Investissement

398

d’Avenir - Program ANR-11-IDEX-0005-02, Laboratoire d’Excellence, INFLAMEX.

AC C

EP

TE D

M AN U

SC

RI PT

393

17

ACCEPTED MANUSCRIPT 399

References

400

1.

401

35. 2.

403 404

Hambly N, Nair P. Monoclonal antibodies for the treatment of refractory asthma. Curr Opin Pulm Med 2014; 20: 87-94.

3.

RI PT

402

Wener RR, Bel EH. Severe refractory asthma: an update. Eur Respir Rev 2013;22:227-

Pavord ID, Korn S, Howarth P, Bleecker ER, Buhl R, Keene ON, et al. Mepolizumab for severe eosinophilic asthma (DREAM): a multicentre, double-blind, placebo-controlled

406

trial. Lancet 2012;380:651-9.

407

4.

SC

405

Miranda C, Busacker A, Balzar S, Trudeau J, Wenzel SE. Distinguishing severe asthma phenotypes: role of age at onset and eosinophilic inflammation. J Allergy Clin Immunol

409

2004;113:101-8. 5.

411 412

Med 2012;18:716-25. 6.

413 414

Wenzel S. Asthma phenotypes: the evolution from clinical to molecular approaches. Nat

Chung KF, Adcock IM. How variability in clinical phenotypes should guide research into disease mechanisms in asthma. Ann Am Thorac Soc 2013;10 Suppl:S109-17.

7.

415

TE D

410

M AN U

408

Baraldo S, Turato G, Bazzan E, Ballarin A, Damin M, Balestro E, et al. Noneosinophilic asthma in children: relation with airway remodelling. Eur Respir J 2011; 38:575-583.

8.

Hirota N, Martin JG. Mechanisms of airway remodelling. Chest 2013;144:1026-32.

417

9.

Brightling CE, Gupta S, Gonem S, Siddiqui S. Lung damage and airway remodelling in

419 420

severe asthma. Clin Exp Allergy 2012;42:638-49.

AC C

418

EP

416

10. Cox G, Thomson NC, Rubin AS, Niven RM, Corris PA, Siersted HC et al. Asthma control during the year after BT. N Engl J Med 2007; 356: 1327-37.

421

11. Pavord ID, Cox G, Thomson NC, Rubin AS, Corris PA, Niven RM et al. Safety and

422

efficacy of BT in symptomatic, severe asthma. Am J Respir Crit Care Med 2007; 176:

423

1185-91.

424

12. Castro M, Rubin AS, Laviolette M, Fiterman J, De Andrade Lima M, Shah PL, et al.

425

Effectiveness and safety of BT in the treatment of severe asthma: a multicenter,

18

ACCEPTED MANUSCRIPT 426

randomized, double-blind, sham-controlled clinical trial. Am J Respir Crit Care Med

427

2010; 181: 116-24.

428 429

13. Wahidi MW, Kraft M. BT for severe asthma. Am J Respir Crit Car Med 2012; 185: 70914. 14. Dombret MC, Alagha K, Boulet L, Brillet PY, Joos G, Laviolette M, et al. BT: a new

431

therapeutic option for the treatment of severe, uncontrolled asthma in adults. Eur Respir

432

Rev 2014; 134: 510-8.

RI PT

430

15. Pretolani M, Dombret MC, Thabut G, Knap D, Hamidi F, Debray MP, et al. Reduction of

434

ASM mass by BT in patients with severe asthma. Am J Respir Crit Care Med 2014;

435

190:1452-4.

SC

433

16. Chung KF, Wenzel SE, Brozek JL, Bush A, Castro M, Sterk PJ, et al. International

437

ERS/ATS guidelines on definition, evaluation and treatment of severe asthma. Eur

438

Respir J 2014;43:343-73.

440

17. Juniper EF, O’Byrne PM, Guyatt GH, Ferrie PJ, King DR. Development and validation of a questionnaire to measure asthma control. Eur Respir J 1999;14:902-7.

TE D

439

M AN U

436

18. Juniper EF, Svensson K, Mork AC, Stahl E. Measurement properties and interpretation

442

of three shortened versions of the asthma control questionnaire. Respir Med 2005;

443

99:553-58.

445

19. Juniper EF, Guyatt GH, Ferrie PJ, Griffith LE. Measuring quality of life in asthma. Am Rev Respir Dis 1993; 147:832-38.

AC C

444

EP

441

446

20. Benayoun L, Druilhe A, Dombret MC, Aubier M, Pretolani M. Airway structural alterations

447

selectively associated with severe asthma. Am J Respir Crit Care Med 2003; 167:1360-

448

8.

449

21. Bergqvist A, Andersson CK, Mori M, Walls AF, Bjermer L, Erjefält JS. Alveolar T-helper

450

type-2 immunity in atopic asthma is associated with poor clinical control. Clin Sci (Lond)

451

2015; 128: 47-56.

452

22. Mori M, Andersson CK, Svedberg KA, Glader P, Bergqvist A, Shikhagaie M et al.

453

Appearance of remodeled and dendritic cell-rich alveolar-lymphoid interfaces provides a

19

ACCEPTED MANUSCRIPT 454

structural basis for increased alveolar antigen uptake in chronic obstructive pulmonary

455

disease. Thorax 2013; 68: 521-31. 23. Bergqvist A, Andersson CK, Hoffmann HJ, Mori M, Shikhagaie M, Krohn IK et al. Marked

457

epithelial cell pathology and leukocyte paucity in persistently symptomatic severe

458

asthma. Am J Respir Crit Care Med 2013; 188: 1475-7.

RI PT

456

459

24. Pavord ID, Thomson NC, Niven RM, Corris PA, Chung KF, Cox G et al. Safety of BT in

460

patients with severe refractory asthma. Research in Severe Asthma Trial Study Group.

461

Ann Allergy Asthma Immunol. 2013;111:402-7.

25. Wechsler ME, Laviolette M, Rubin AS, Fiterman J, Lapa e Silva JR, Shah PL, Fiss E,

463

Olivenstein R, Thomson NC, Niven RM, Pavord ID, Simoff M, Hales JB, McEvoy C,

464

Slebos DJ, Holmes M, Phillips MJ, Erzurum SC, Hanania NA, Sumino K, Kraft M, Cox G,

465

Sterman DH, Hogarth K, Kline JN, Mansur AH, Louie BE, Leeds WM, Barbers RG,

466

Austin JH, Shargill NS, Quiring J, Armstrong B, Castro M; Asthma Intervention Research

467

2 Trial Study Group. Bronchial thermoplasty: Long-term safety and effectiveness in

468

patients with severe persistent asthma. J Allergy Clin Immunol. 2013;132:1295-302.

469

26. Chakir J, Haj-Salem I, Gras D, Joubert P, Beaudoin ÈL, Biardel S, et al. Effects of

470

Bronchial Thermoplasty on Airway Smooth Muscle and Collagen Deposition in Asthma.

471

Ann Am Thorac Soc. 2015;12:1612-18.

EP

TE D

M AN U

SC

462

27. Danek CJ, Lombard CM, Dungworth DL, Cox PG, Miller JD, Biggs MJ et al. Reduction in

473

airway hyperresponsiveness to methacholine by the application of RF energy in dogs. J

474

Appl Physiol 2004;97:1946-53

475 476 477 478

AC C

472

28. Miller JD, Cox G, Vincic L, Lombard CM, Loomas BE, Danek CJ. A prospective feasibility study of BT in the human airway. Chest 2005;127:1999-2006. 29. Bourdin A, Neveu D, Vachier I, Paganin F, Godard P, Chanez P. Specificity of basement membrane thickening in severe asthma. J Allergy Clin Immunol 2007;119:1367-74.

479

30. Vrugt B, Wilson S, Bron A, Holgate ST, Djukanovic R, Aalbers R. Bronchial

480

angiogenesis in severe glucocorticoid-dependent asthma. Eur Respir J. 2000;15:1014-2.

481

31. Lee LY, Yu J. Sensory nerves in lung and airways. Compr Physiol. 2014; 4:287-324.

20

ACCEPTED MANUSCRIPT 482

32. van der Velden VH, Hulsmann AR. Autonomic innervation of human airways: structure,

483

function, and pathophysiology in asthma. Neuroimmunomodulation. 1999;6:145-59.

484

33. Linnoila RI. Functional facets of the pulmonary neuroendocrine system. Lab Invest.

485

2006;86:425-44. 34. Weichselbaum M, Sparrow MP, Hamilton EJ, Thompson PJ, Knight DA. A confocal

487

microscopic study of solitary pulmonary neuroendocrine cells in human airway

488

epithelium. Respir Res 2005;6:115-126.

490

35. Adriaensen D, Timmermans JP. Purinergic signalling in the lung: important in asthma and COPD? Curr Opin Pharmacol 2004;4:207-14.

SC

489

RI PT

486

36. Darcy RD, Doeing DC, Hogarth DK, Dugan K, Naureckas ET, White SR. Airway

492

inflammation after bronchial thermoplasty for severe asthma. Ann Am Thorac Soc

493

2015;12:1302-09.

AC C

EP

TE D

M AN U

491

21

ACCEPTED MANUSCRIPT Figure legends

495

Figure 1. Histopathological alterations induced by BT. Bronchial biopsies were assessed

496

for immunohistochemical and morphometric analyses of ASM area (as % of positive α-actin

497

immunostained area over the submucosal tissue area) (A), SBM thickening (in µm) (B),

498

submucosal blood (C) and lymphatic vessels (D), submucosal nerve fibers (E), after

499

immunolabelling of CD31, D2-40 and PGP9.5 antigens, respectively (all expressed as ‰ of

500

positive immunostaining over the submucosal tissue area), of PGP9.5+ nerves infiltrating the

501

ASM (in numbers per mm2 of ASM) (F), of mucous glands (as % prevalence) (G), of

502

eosinophils (H) and neutrophils (I), as proportions of bronchial mucosa with positive

503

immunoreactivity for ECP and MPO, respectively (in %), of regenerating epithelium (as %

504

prevalence) (J), of stratified columnar epithelium (as % prevalence) (K), metaplastic

505

epithelium (as % prevalence) (L), squamous metaplastic epithelium (as % prevalence) (M),

506

goblet cell hypertrophy/ hyperplasia (as % prevalence) (N), and of epithelium neuroendocrine

507

(PGP9.5-positive) cells (in numbers per mm2 of intact areas of bronchial epithelium) (O).

508

Comparisons were made by Student’s t test for paired values.

509

Figure 2. Structural effects of BT in bronchial biopsy specimens from severe

510

asthmatics. Bright field micrographs of bronchial biopsies subjected to quadruple

511

immunohistochemical staining for smooth muscle actin (red), the vascular endothelial marker

512

CD31 (green), lymph endothelial marker, podoplanin (D2-40, brown), and the neuronal

513

marker PGP.9.5 (black). A and B illustrate biopsies taken before and 3 months after BT,

514

respectively (note that B, rather than representing the average, exemplifies a case where

515

smooth muscle was virtually absent). C illustrates a neuroendocrine cells (NEC) in the

516

bronchial epithelium detected by the nuclear distribution of PGP (arrowhead). Sub-epithelial

517

nerves (arrowheads) in the sub-epithelial region before (D) and after (E) BT. Smooth muscle-

518

associated nerves (arrowhead) are exemplified in F in a biopsy collected before BT

519

treatment. Scale bars: 250 µm (A,B), 40 µm (C,D,E,F). sm, bv and lv, denote smooth

520

muscle, blood vessels and lymphatic vessels, respectively.

AC C

EP

TE D

M AN U

SC

RI PT

494

22

ACCEPTED MANUSCRIPT

Figure 3. Differences in histopathological alterations in BT- responsive and -partially

522

responsive severe asthmatics.

523

ASM area (A), SBM thickening (B), sub-mucosal PGP+ nerves (C), ASM-associated PGP+

524

nerves (D) and neuroendocrine epithelial PGP+ cells (E) in bronchial biopsies from severe

525

asthmatics collected before (n = 15) and 3 months after BT. Patients were separated into two

526

groups, according to their clinical responsiveness (n = 11, score of ACT ≥ 15), or partial, or

527

no responsiveness (n = 4, score of ACT < 15) to BT at 12 months.

528

Overall P values were of < 0.01 (A), 0.31 (B), < 0.01 (C), 0.06 (D), and < 0.01 (E) (Kruskal-

529

Wallis test). * P < 0.05, as compared to values obtained before BT; † P < 0.05, as compared

530

to patients responsive to BT (Student’s t test for paired values).

SC

RI PT

521

AC C

EP

TE D

M AN U

531 532 533

ACCEPTED MANUSCRIPT

RI PT

se Table 1: Characteristics of severe asthmatics before and after BT Parameter

Before BT

3 months after BT

12 months after BT

No. of patients

15

15

15 ___

___

P value

7 (53)

Age – years

46.9 ± 11.9

___

___

___

13/1/1

___

___

___

28.2 ± 4.6

___

___

___

21.3 ± 18.2

___

___

___

24.2 ± 14.5

___

___

___

10 (67)

___

___

___

Smoking history – no. of never smokers/ex-smokers/active smokers 2

Age of asthma onset – years Asthma duration – years With history of atopy – no. (%) 9

Blood eosinophils count – 10 /L Total serum IgE – International Units/mL Respiratory function

140 (110-231)

b

141 (102-222)

290 (135-615)

b

358 (94-545)

b

b

158 (135-215)

b

0.34

231 (120-316)

b

0.92

2068 ± 682

1990 ± 480

2100 ± 680

0.66

Post-bronchodilator FEV1 – mL

2250 ± 656

2300 ± 820

2250 ± 670

0.93

Pre-bronchodilator FEV1 – % of predicted

67.1 ± 19.5

63.5 ± 13.4

66.6 ± 16.8

0.65

Post-bronchodilator FEV1 – % of predicted

71.0 ± 16.6

70.2 ± 13.2

70.8 ± 15.7

0.97

Pre-bronchodilator FVC – mL

3274 ± 833

3190 ± 710

3400 ± 980

0.43

Post-bronchodilator FVC – mL

3450 ± 798

3320 ± 740

3450 ± 980

0.45

Pre-bronchodilator FEV1/FVC – % of predicted

64.3 ± 12.3

61.9 ± 12.8

60.6 ± 8.3

0.18

Post-bronchodilator FEV1/FVC – % of predicted

65.3 ± 13.2

65.1 ± 11.9

68.0 ± 12.0

0.47

Reversibility to β2-agonists – mL

19.3 ± 20.3

32.4 ± 67.3

15.2 ± 14.8

0.42

Reversibility to β2-agonists – %

6.1 ± 6.0

7.4 ± 10.6

4.7 ± 4.3

0.55

AC C

EP

TE D

Pre-bronchodilator FEV1 – mL

M AN U

Body Mass Index – kg per m

SC

Women – no. (%)

___

b

a

ACCEPTED MANUSCRIPT

Treatments On long-acting β2-agonists – no.

15

Dose of ICS – µg of equivalents of beclomethasone per day

RI PT

Table 1 (continued)

15

2133 ± 516

2000 ± 0

2000 ± 0

0.38

On maintenance use of OCS – no. (%)

10 (67)

9 (60)

8 (53)

0.15

Dose of oral prednisone – mg per day

31.5 ± 11.1

20.6 ± 12.4

13.8 ± 5.2 *

0.002

10 (67)

0

0

8 (53)

6 (40)

7 (47)

0.57

7 (47)

6 (40)

7 (60) *

0.04

On nebulized anti-cholinergics and β2-agonists – no. (%)

9 (60)

4 (27) *

3 (20) *

< 0.001

On theophylline – no. (%)

2 (13)

4 (7)

1 (7)

15 (100)

6 (40) *

4 (27) *

< 0.001

8.5 ± 2.8

15.7 ± 4.8 *

16.4 ± 6.9 *

< 0.001

2.6 ± 0.9

3.7 ± 1.5

4.2 ± 1.5 *

< 0.001

Annual rate of severe exacerbations

9.7 ± 1.3

0.7 ± 0.3 *

0.7 ± 0.4 *

< 0.001

Annual rate of hospitalization for asthma

1.7 ± 0.8

0.2 ± 0.1 *

0.2 ± 0.1 *

< 0.001

Annual rate of visits to emergency department

3.3 ± 1.0

0.5 ± 0.3 *

0.3 ± 0.2 *

< 0.001

Annual rate of hospitalization in ICU

0.9 ± 0.7

0.2 ± 0.1 *

0.1 ± 0.1 *

0.02

On anti-IgE – no. (%)

SC

15

c

On anti-histamine – no. (%)

Asthma control With uncontrolled asthma – no. (%) Score on ACT

EP

TE D

Score on AQLQ

M AN U

On anti-leukotrienes – no. (%)

c

< 0.001

0.15

AC C

Abbreviations: BT, bronchial thermoplasty; FEV1 = Forced Expiratory Volume in One second; FVC = Forced Vital Capacity; ICS = inhaled corticosteroids; OCS, oral corticosteroids; ACT = asthma control test; AQLQ = asthma quality of life questionnaire; ICU, ICU, intensive care unit. Data are n (%), or means ± SD, or medians (interquartile range), or means ± SE for annual rates of severe exacerbations, hospitalizations for asthma, visits to emergency department and hospitalizations in ICU a One-way ANOVA b log-transformed values of this variable were used for statistical testing. c Treatment with anti-IgE was stopped at least 6 months before the first BT session Bold denotes statistical significance

* P < 0.05, as compared to values obtained before BT (Student’s t test, or Fisher exact test, two-tailed)

ACCEPTED MANUSCRIPT

Table 2. Correlation analyses between histopathological parameters and asthma control 3 and 12 months after BT

Score on ACT ra

95%CI

Score on AQLQ

No. of severe exacerbations

No. of visits to emergency department

No. of hospitalization for asthma

No. of hospitalization in ICU

r

95%CI

r

95%CI

r

95%CI

r

95%CI

r

95%CI

0.690 #

0.30/1.00

0.616 †

0.21/1.00

0.457 *

0.04/0.87

0.309

-0.10/0.72

0.372

-0.04/0.79

0.281

-0.13/0.69

0.408

-0.02/0.83

0.345

-0.07/0.76

0.689 #

0.30/1.00

0.400

-0.02/0.82

0.168

-0.25/0.59

- 0.090

-0.53/0.35

- 0.052

-0.48/037

Results 3 months after bronchial thermoplasty

RI PT

Parameter

- 0.600 † -1.00/-0.20

- 0.321

-0.73/0.08

SBM thickening (µm)

- 0.503 * -0.93/-0.08

- 0.332

-0.74/0.08

Sub-mucosal nerves (‰ PGP immunoreactivity)

- 0.367

-0.78/0.04

- 0.185

-0.61/0.24

ASM-associated PGP+ nerves (pixels per mm2 ASM)

- 0.187

-0.61/0.28

- 0.041

-0.42/0.33

0.433 *

0.00/0.87

0.219

-0.21/0.65

0.154

-0.25/0.56

Number of PGP+ neuroendocrine cells per mm2 epithelium

- 0.508 * -0.94/-0.08

- 0.376

-0.80/0.04

0.526 †

0.13/0.93

0.510 *

0.08/0.94

0.592 †

0.20/0.98

0.423 *

0.02/0.83

ASM (% of submucosal area)

- 0.516 * -0.94/-0.09

- 0.432 * -0.86/0.00

0.580 †

0.17/0.99

0.572 †

0.16/0.98

0.310

-0.11/0.73

0.189

-0.22/0.60

SBM thickening (µm)

- 0.388

-0.81/0.03

- 0.364

-0.80/0.07

0.341

-0.09/0.77

0.296

-0.12/0.71

0.386

-0.05/0.82

0.255

-0.17/0.68

Sub-mucosal nerves (‰ PGP immunoreactivity)

- 0.236

-0.66/0.18

- 0.232

-0.64/0.18

0.667 †

0.19/1.00

0.518 *

0.08/0.95

0.202

-0.22/0.63

0.001

-0.08/0.08

ASM-associated PGP+ nerves (pixels per mm2 ASM)

- 0.144

-0.57/0.28

- 0.112

-0.54/0.32

0.351

-0.08/0.78

0.231

-0.18/0.64

0.092

-0.33/0.52

- 0.034

-0.48/0.41

Number of PGP+ neuroendocrine cells per mm2 epithelium

- 0.502 * -0.92/-0.08

0.506 *

0.08/0.93

0.538 *

0.08/0.99

0.501 *

0.08/0.92

0.387

-0.05/0.82

EP

M AN U

TE D

Results 12 months after bronchial thermoplasty

SC

ASM (% of submucosal area)

- 0.452 * -0.88/-0.02

AC C

Abbreviations: BT, bronchial thermoplasty; ACT, asthma control test; AQLQ, asthma quality of life questionnaire; ICU, intensive care unit; CI, confidence interval; ASM, airway smooth muscle; SBM, subepithelial basement membrane; PGP, Protein Gene Product (nerve and epithelium neuroendocrine cell marker). a Spearman’s rank-order method and Benjamini and Hochberg correction * P < 0.05; † P < 0.01; # P < 0.001

ACCEPTED MANUSCRIPT

B

20 10

0

After BT

E 10 8 6 4 2 0

Before BT

After BT

20

0

Before BT

40 20 0

AC C

80

Before BT

After BT

M 100

P = 0.55

80 60 40 20 0

Before BT

After BT

Stratified columnar epithelium (%)

P = 0.24

100

60

After BT

EP

J Regenerating epithelium(%)

3 2 1 0

Before BT

40 20 0

Before BT

3000

After BT

P = 0.02

2000

4 3 2

P = 0.28

1 0

K

100

Before BT

After BT

P = 0.86

80 60 40 20 0

Before BT

Before BT

After BT

I 20

P = 0.85

15 10 5 0

L 100

Before BT

After BT

P = 0.34

80 60 40 20 0

Before BT

After BT

O P = 0.40

80 60 40 20 0

0

After BT

N 100

1000

After BT

TE D

40

Goblet cell hypertrophy/hyperplasia (%)

Mucous glands (%)

60

60

F

4

H P = 0.57

After BT

P < 0.001

5

Eosinophils (% immunoreactivity)

G

Before BT

80

SC

P = 0.22 Submucosal PGP+ nerves (‰)

Submucosal lymphatic vessels (‰)

D

Squamous metaplastic epithelium (%)

2

Metaplastic epithelium (%)

Before BT

4

PGP+ neuroendocrine cells per mm2 epithelium (%)

0

6

P = 0.21

100

RI PT

30

C Submucosal blood vessels (‰)

SBM thickening (µm)

40

P = 0.02

8

ASM-associated PGP+ nerves (pixels per mm2 ASM)

P < 0.001

M AN U

ASM area (% of submucosal area)

A

Neutrophils (% immunoreactivity)

Figure 1

Before BT

After BT

60

P = 0.02

40

20

0

Before BT

After BT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Figure 3 B

RI PT

4 2 0

D

2

*

1

*

2000

AC C

3

0

3000

EP

4

TE D

0

*

1000

*

0

E Neuroendocrine PGP+ cells/mm2 epithelium

10

6

SC

*

Responsive to bronchial thermoplasty Partially responsive to bronchial thermoplasty

M AN U

†

30 20

SBM thickening (µm)

40

C Sub-mucosal PGP+ nerves (‰)

Before bronchial thermoplasty

8

ASM-associated PGP+ nerves (pixels/mm2 ASM)

ASM area (% of sumucosal area)

A

60

40

20

0

†

*

ACCEPTED MANUSCRIPT 1 2 3

Effectiveness of bronchial thermoplasty in patients with severe

5

refractory asthma: clinical and histopathological correlations

RI PT

4

6

Marina Pretolani, Pharm. D., Ph.D., Anders Bergqvist, Ph.D., Marie-Christine Dombret,

9

M.D, Gabriel Thabut, M.D., Ph.D., Dominique Knapp, Fatima Hamidi, MS,

10

Loubna Alavoine, M.D., Camille Taillé, M.D., Ph.D., Pascal Chanez, M.D., Ph.D.,

11

Jonas Erjefält, M.D., Ph.D., and Michel Aubier, M.D.

M AN U

SC

7 8

12 13 14 15 16 17 18

AC C

EP

TE D

Online Repository

1

ACCEPTED MANUSCRIPT Methods

20

Patients

21

All patients had a follow up in our asthma clinic for at least 12 months and belong to the

22

French national asthma cohort COBRA (Cohort Bronchial Obstruction and Asthma). The

23

later is a prospective, multicenter cohort with a 10 years follow-up and a frequency visit every

24

6 months. All baseline clinical outcomes were extracted from the COBRA database during

25

the year preceding the inclusion in the present study. Table 1 in the manuscript describes the

26

characteristics of the 15 severe asthmatics examined at inclusion (i.e., 15 days before the

27

first bronchial thermoplasty [BT] session) and 3 and 12 months after BT. The majority of

28

these patients were female, had history of atopy, with mild blood eosinophilia and total serum

29

IgE. All patients experienced symptomatic uncontrolled severe asthma, with mean values of

30

asthma control test (ACT) and asthma quality of life questionnaire (AQLQ) scores of 8.47

31

and 2.56, respectively, despite high-dose of inhaled corticosteroids (ICS, ≥ 1000 µg

32

fluticasone propionate per day) and long-lasting β2 agonists (LABA). They also had more

33

than 3 clinically significant exacerbations (mean = 9.7), 2 to 12 hospitalizations for asthma

34

(mean = 1.7) and in intensive care unit (ICU, mean = 0.9) and 1 to 12 visits to emergency

35

department (mean = 3.3) in the previous year. These severe asthmatics showed airflow

36

obstruction, with mean values of pre- and post-bronchodilator forced expiratory volume in

37

one second (FEV1) < 80% predicted and of pre- and post-bronchodilator FEV1/forced vital

38

capacity (FVC) < 70%. Ten patients required maintenance with OCS, with mean daily doses

39

of 31.5 mg prednisone. At the time of inclusion, they also received add-on therapies,

40

including anti-leukotrienes (8 out of 15 patients), anti-histamine (7 out of 15 patients),

41

nebulized anti-cholinergics (9 out of 15 patients) and theophylline (2 out of 15 patients). Ten

42

patients met the criteria for receiving omalizumab for 6 months during the year before entry

43

in the protocol, without successful clinical outcome. Omalizumab was stopped at least 6

44

months before the first BT session.

AC C

EP

TE D

M AN U

SC

RI PT

19

2

ACCEPTED MANUSCRIPT 45

Patients were managed according to the best standard care in tertiary teaching outpatient

46

clinic of the Bichat hospital.

47 Biopsy collection

49

All biopsies were collected at all lobes airway carinas. In all patients, 3 biopsies were

50

collected at the right lower lobe airway carinas B7-B8, B8-B9, B9-B10 three months before

51

they were included in the protocol. These biopsies are routinely performed in our clinic

52

department in patients with severe asthma at the time of their inclusion in the COBRA cohort.

53

To assess the variability within biopsies, we calculated the intraclass correlation coefficient

54

(ICC), as an index of repeatability. To this end, we used the 3 biopsies collected at the right

55

lower lobe airway carinas (B7-B8, B8-B9, B9-B10) 3 months before the patients were

56

included in the protocol. The ICC between biopsies within patients was 0.46.

57

BT procedure

58

The same operator (M-C.D.) was in charge of all BT procedures. On top of local anesthesia

59

administered by the endoscopist and after hydroxyzine and atropine oral premedication, all

60

patients received a titrated remifentanil target controlled infusion, associated, if required, to

61

small propofol boluses (5 to 10 mg). All the procedures were performed in the sitting position.

62

Fifty mg of oral prednisone were administered two days before and three days after each of

63

the 3 BT sessions.

64

Upon BT treatment, patients experienced an increase in respiratory adverse events related

65

to asthma, cough, wheeze, expectoration, dyspnea and nocturnal awakening. The majority of

66

respiratory adverse events occurred within 1 day of the bronchoscopy and resolved within 7

67

days. On the day, or the day after BT, 3 subjects required hospitalization for severe

68

exacerbations of asthma, which resolved within 7 days with usual standard therapy. All the

69

other patients were discharged from the hospital 24 hours after the procedure. One patient

70

had hemoptysis 2 weeks after the second session, which required arterial embolization. This

71

event resolves rapidly and the patient underwent the third BT session without any adverse

AC C

EP

TE D

M AN U

SC

RI PT

48

3

ACCEPTED MANUSCRIPT event.

73

No adverse event related to BT was noted during the one-year follow-up.

74

Immunohistochemistry and morphometry

75

Primary antibodies and experimental conditions for their use are described in Table E1.

76

Quadruple immunohistochemical staining protocol

77

Slides were first blocked using a dual endogenous enzyme-blocking reagent (Dako,

78

Glostrup, Denmark) in order to quench endogenous peroxidase and alkaline phosphatase.

79

Next, slides were incubated with 10% goat serum (Dako) for 20 minutes to block unspecific

80

binding of the secondary antibodies made in goat. Slides were then incubated with an anti-

81

PGP9.5 antibody during 60 minutes, followed by incubation with a dextran polymer chain

82

containing peroxidase molecules and secondary antibodies against mouse immunoglobulins

83

(Dako) for 30 minutes. Next, Deep Space Black Chromogen (Biocare Medical, Göteborg,

84

Sweden) was added for 10 minutes to stain PGP9.5 in black. Thereafter, sections were

85

incubated for 5 minutes with a blocking solution (Biocare Medical) in order to denature

86

residual secondary antibodies on the dextran polymer chain and to avoid cross-reactivity.

87

Slides were then incubated with the lymph endothelial marker antibody, DP-40, during 60

88

minutes, followed by incubation with a horseradish peroxidase (HRP) conjugated anti-mouse

89

antibody for 30 minutes. After this, 3’3 diaminobenzide (DAB) chromogen (Dako) was added

90

to the slides for 10 minutes to stain D2-40 in brown. Next, alkaline phosphatase conjugated

91

antibodies against smooth muscle actin were added to the slides for 60 minutes, followed by

92

incubation with Permanent Red Chromogen (Dako) for 10 minutes to stain smooth muscle

93

actin in red. Finally, to stain blood vessels, slides were incubated with an anti-CD31 antibody

94

for 60 minutes, followed by incubation with HRP-conjugated anti-mouse antibodies for 30

95

minutes. Slides were then incubated with Vina Green Chromogen (Biocare Medical) for 10

96

minutes twice to stain CD31 in green. Background staining was visualized with hematoxylin

97

and sections were placed in a xylene bath before mounting with Pertex mounting medium

98

(Histolab Products, Göteborg, Sweden).

AC C

EP

TE D

M AN U

SC

RI PT

72

4

ACCEPTED MANUSCRIPT The present immunohistochemical protocol resulted in simultaneously detection of PGP9.5

100

positive pulmonary neuroendocrine cell (block epithelial nuclei) and nerve fibers (black

101

nerves), DAB brown D2-40 positive lymph vessels, PR red airway smooth muscle and Vina

102

green CD31 positive blood vessels (blood vessels were also separated from e.g. lymph

103

vessels due to their smooth muscle content (exemplified in Fig E1). Cross-reactivity was

104

avoided by a combination of steric hindrance, by denaturing residual secondary antibodies,

105

and by using directly enzyme-conjugated antibodies. After PGP staining, the secondary

106

antibodies on the dextran chain were inactivated by denaturing solution (DNS001L, Biocare

107

Medical), thereby preventing any further binding of primary or secondary antibodies in the

108

subsequent protocol steps. Staining of lymph vessels was saturated by insoluble DAB

109

precipitate, to achieve a steric hindrance of further HRP-based chromogen (in this case Vina

110

Green Chromogen, BRR807A). Antibodies against alpha smooth muscle actin were directly

111

conjugated with AP, and due to the use of denaturing solution, unable to bind to the dextran

112

chain used in the detection of PGP positive cells or nerve fibers. As exemplified from Fig E1,

113

the resulting quadruple staining yielded a distinct and clear color separation of the immune

114

stained structures.

115

No staining was observed when omitting the primary antibodies, or replacing them with the

116

corresponding control isotypes.

117

Double staining protocol

118

Slides were first incubated with 0.3% hydrogen peroxidase for 10 minutes to quench

119

endogenous peroxidase. Next, slides were incubated with 10% goat serum (Dako) for 20

120

minutes to block unspecific binding of the secondary antibodies made in goat. Slides were

121

then incubated with a mouse monoclonal anti-eosinophil cationic protein (ECP) antibody

122

(eosinophil marker) for 60 minutes, followed by incubation with horseradish peroxidase-

123

conjugated anti-mouse antibodies for 30 minutes. After this procedure, DAB chromogen

124

(Dako) was added to the slides for 10 minutes to stain ECP in brown. Next, slides were

125

incubated with a rabbit polyclonal anti-myeloperoxidase (MPO) antibody (neutrophil marker)

AC C

EP

TE D

M AN U

SC

RI PT

99

5

ACCEPTED MANUSCRIPT for 60 minutes, followed by incubation with horseradish peroxidase-conjugated anti-rabbit

127

antibodies for 30 minutes. Slides were then incubated with Vina Green Chromogen (Biocare

128

Medical) for 10 minutes to stain MPO in green. Background staining was visualized with

129

haematoxylin and sections were placed in a xylene bath before mounting with Pertex

130

mounting medium (Histolab Products).

131

Morphological assessment and staining quantification

132

After robotized immunohistochemistry, all stained slides (n=600) were digitally scanned in an

133

automated digital slide-scanner (Scanscope CS, Aperio Technologies, Leica Microsystems,

134

Buffalo Grove, IL, USA) operating with a 40 x microscope lens. ASM area was determined

135

with the ImageScope software (v. 10.0.36.1805; Aperio Technologies) by manually

136

delineating ASM surfaces, under guidance by α-smooth-muscle actin immunostaining

137

(E1,E2). In each biopsy, the area of ASM was normalized to the area of sub-mucosa, which

138

was obtained by manually excluding glands, cartilage and artifacts. Briefly, the program

139

measures and automatically converts the total number of pixels to a metric area unit (in mm2)

140

within a region of interest. Using this approach, the area of ASM and sub-mucosa could be

141

established in a rapid and reproducible approach. As shown in Fig E2A, there was a trend,

142

although not significant, towards a decrease in the total biopsy area 3 months after BT.

143

However, small biopsies (less than 0.150 mm2) were excluded since they were considered to

144

be too small for analysis. In addition, epithelium area was very similar before and 3 months

145

after BT (Fig E2B). However, biopsies that contained little intact epithelium (defined as less

146

than 250 µm of combined length) were excluded when the number of neuroendocrine cells

147

(NEC) was analyzed (see below).

148

To determine the stability of histological measurements over time in the absence of BT, we

149

collected 2 biopsies in the right lobe (B8-B9) of the 15 severe asthmatics approximately 3

150

months before their entry in the protocol and we compared ASM area to that obtained in the

151

same locations 15 days before the 1st of the 3 BT sessions. The extent of ASM area was

152

comparable at these two time-points (median [25-75] interquartile range, 19.5 [15.8-24.8] and

AC C

EP

TE D

M AN U

SC

RI PT

126

6

ACCEPTED MANUSCRIPT 16.2 [12.0-21.4], 3 months and 15 days before BT, respectively; P = 0.30, Fig E3A). In

154

parallel, we determined the repeatability of the measures of PGP+ nerves in the bronchial

155

sub-mucosa and within the ASM, since these parameters were also targeted by BT. Although

156

with some variations, the overall comparisons between measures performed in the 3

157

biopsies obtained 3 months before entry in the BT protocol and those collected in 2 weeks

158

before the first BT session at the same sites (i.e., the right lower lobe) were not statistically

159

different (P = 0.53, for mucosal PGP+ nerves and P = 0.35, for ASM-associated PGP+

160

nerves) (Fig E3C and D).

161

Finally, the extent of ASM area measured 2 weeks before the first BT session in the 3

162

biopsies collected in the right lower lobe, was not statistically different than that evaluated in

163

the biopsies obtained from the remaining 4 lobes (P = 0.12, Fig E3B). This finding suggests

164

that the biopsies performed before BT did not affect the subsequent biopsies performed 3

165

months after the last BT session.

166

Sub-epithelial basement membrane (SBM) thickening was determined by multiple point-to-

167

point measurements using ImageScope (Aperio technologies) with approximately 50-µm

168

intervals between each measurement (corresponding approximately to one measurement

169

per every 7th epithelium basal cell). Automated quantification of immune staining was carried

170

out using Visiomorph DP (Visiopharm, Hørsholm, Denmark) by employing positive staining

171

recognizing algorithms. Briefly, by dividing positively stained area (i.e., total amount of

172

immunoreactive pixels) with the total tissue area (pixels), the percentage of positive staining

173

was calculated for each marker. Immuno-reactivity for the vascular endothelial marker,

174

CD31, for lymphatic vessels, D2-40, and for the panaxonal marker, Protein Gene Product

175

(PGP) 9.5 (marker of nerve fibers), was determined in the bronchial sub-mucosa, after

176

excluding ASM, glands and cartilage by manual delineation. For sub-mucosal tissue

177

analysis, a usable piece of tissue was defined as presence of well-preserved lamina propria.

178

The average number of pre-BT and post-BT biopsies analyzed per patient for ASM was 9.7

179

and 9.3 respectively. PGP9.5 within, or in close vicinity (distance 0 to 3 µm) to ASM, was

180

determined by dividing the number of positively stained pixels obtained in Visiomorph DP

AC C

EP

TE D

M AN U

SC

RI PT

153

7

ACCEPTED MANUSCRIPT 181

(Visiopharm) with the ASM tissue area determined previously in ImageScope (Aperio

182

Technologies), yielding a value of pixels per mm . NEC, defined as characteristically

183

PGP9.5-positive intraepithelial cells, were manually counted in stretches of intact epithelium,

184

using a threshold value of at least 250 µm of combined length.

185

The surface area of intact epithelium was assessed in ImageScope (Aperio Technologies,

186

US) to obtain a value of cells per mm2. The average number of pre-BT and post-BT biopsies

187

analyzed per patient for NEC were 5.6 and 6.0 respectively. ECP and MPO immuno-

188

reactivity was determined as described above, in Visiomorph DP (Visiopharm) in the whole

189

biopsy sections, after excluding glands and cartilage by manual delineation. Morphological

190

characteristics of the bronchial epithelium (stratified columnar epithelium, metaplastic

191

epithelium, squamous metaplastic epithelium, goblet cell hyperplasia/ hypertrophy and

192

regenerated epithelium were assessed qualitatively on hematoxylin stained sections using

193

ImageScope (Aperio Technologies), as previously described (E3). The presence of glands

194

was also qualitatively assessed on haematoxylin-stained tissue sections. All the parameters

195

were examined in two non-contiguous sections from the same biopsy, at different depths for

196

each biopsy, and in sections from 10 different biopsies. Sections were randomly assigned

197

and analyzed by one observer who was unaware about the clinical group assignment. The

198

final figure was the median of all the measurements obtained for each patient. For

199

neuroendocrine cells, the mean value was used as the final figure. Values were obtained by

200

considering mean values from the coefficients of variance obtained in each patient.

201

Finally, collagen was evidenced using Masson’s trichrome dye and the proportion of biopsy

202

area stained for collagen was quantified by image analysis (E4, E5). Briefly, digital images of

203

Masson’s trichrome-stained tissue slides were examined at x 20 magnification, using a slide

204

scanner, coupled to an image analyzer (Calopix®, TRIBVN, Châtillon, France). The

205

proportion of collagen-positive biopsy area was assessed using an interval scale (0%, 1-

206

20%, 21-40%, 41-60%, 61-80%, and 81-100%). In addition, the intensity of collagen staining

207

was determined using an immuno-surface algorithm, that defined four categories of

208

magnitude, as follows: absence = score 0 (blue); low = score 1 (yellow); moderate = score 2

AC C

EP

TE D

M AN U

SC

RI PT

2

8

ACCEPTED MANUSCRIPT (orange) and high = score 3 (red) (E4, E5) (Fig E4).

210

All of the above immune-histological and morphometrical examinations were assessed in the

211

10 biopsies collected before BT and values were compared to those obtained in the 4 BT-

212

treated lobes at 3 months. In separated analyses, values obtained at 3 months in biopsy

213

specimens from the middle BT-untreated lung lobe were compared to those collected at the

214

same time in the 8 BT-treated lung sites and to those of the 10 biopsies collected before BT

215

(Table E3).

AC C

EP

TE D

M AN U

SC

RI PT

209

9

ACCEPTED MANUSCRIPT 216

References

217

E1.

Pretolani M, Dombret MC, Thabut G, Knap D, Hamidi F, Debray MP, et al. Reduction of

218

airway smooth muscle mass by bronchial thermoplasty in patients with severe asthma.

219

Am J Respir Crit Care Med 2014; 190:1452-4. E2.

Benayoun L, Druilhe A, Dombret MC, Aubier M, Pretolani M. Airway structural

RI PT

220 221

alterations selectively associated with severe asthma. Am J Respir Crit Care Med

222

2003; 167:1360-8. E3.

Bergqvist A, Andersson CK, Hoffmann HJ, Mori M, Shikhagaie M, Krohn IK et al.

SC

223

Marked epithelial cell pathology and leukocyte paucity in persistently symptomatic

225

severe asthma. Am J Respir Crit Care Med 2013; 188: 1475-7.

226

E4.

M AN U

224

Trudel D, Desmeules P, Turcotte S, Plante M, Grégoire J, Renaud MC, et al. Visual and automated assessment of matrix metalloproteinase-14 tissue expression for the

228

evaluation of ovarian cancer prognosis. Mod Pathol 2014;27:1394-404.

229

E5.

TE D

227

Aubier M, Thabut G, Hamidi F, Guillou N, Brard J, Dombret MC, Borensztajn K, Aitilalne B, Poirier I, Roland-Nicaise P, Taillé C, Pretolani M. Airway smooth muscle

231

enlargement is associated with protease-activated receptor 2/ligand overexpression in

232

patients with difficult-to-control severe asthma. J Allergy Clin Immunol. 2016 Mar 18.

233

[Epub ahead of print].

AC C

EP

230

10

ACCEPTED MANUSCRIPT Legends of Supplementary Figures

235

Figure E1. Micrographs exemplifying simultaneous visualization of multiple bronchial

236

wall structures and granulocyte infiltration pattern. Bright field micrographs of bronchial

237

biopsies subjected to quadruple immune-histochemical staining for smooth muscle actin (sm,

238

red), the vascular endothelial marker CD31 (blood vessel, bv, green), lymph endothelial

239

marker, podoplanin (lv, D2-40, brown), and the neuronal marker PGP.9.5 (n, black). Blood

240

and lymphatic vessels were clearly separated by the present marker combination (A,B),

241

which, apart from vessels, allowed for simultaneous and robust identification of smooth

242

muscle bundles (sm) and nerves (“n” in A). Double staining for eosinophils (brown) and

243

neutrophils (green) was used to assess tissue granulocytes, here exemplified in a biopsy

244

collected 3 months after BT (C). Scale bars: 250 µm (A), 70 µm (B and F), 50 µm (C), 140

245

µm.

246

Figure E2. Total biopsy area and bronchial epithelium area measured before and after

247

BT. Total biopsy area (A) and total surface of the bronchial epithelium (B), both expressed in

248

mm2, were assessed by morphometry on Hematolyin & Eosin-stained bronchial tissue

249

sections obtained by fibroscopy in 15 severe asthmatics 15 days before the 1st of the 3 BT

250

sessions and 3 months after the last BT session. Measurements were made in the 5 lung

251

lobes before BT, and in the 4 BT-treated lung lobes at 3 months. Each dot represents mean

252

value of the different measurements performed in the 5, or 4 lobes. Horizontal red bars

253

denote median values. Statistical significance was calculated using Student’s t test for paired

254

values.

SC

M AN U

TE D

EP

AC C

255

RI PT

234

256

Figure E3. ASM area, mucosal and ASM-associated PGP+ nerves, measured 3 months

257

and 2 weeks before BT. A bronchoscopy was performed in the 15 severe asthmatics 3

258

months before their entry in the protocol and 15 days before the 1st of the 3 BT sessions. (A)

259

ASM area (expressed as % of mucosal area), (C) mucosal PGP+ nerves (expressed in %

260

immunoreactivity) and (D) ASM-associated PGP+ nerves (expressed in pixels/mm2) were

11

ACCEPTED MANUSCRIPT measured by morphometry and image analysis in 2 biopsies obtained in the right lower lobe

262

(RLL) (B8-B9) and mean values were calculated. Horizontal red bars represent median

263

values. In addition, values of ASM area were compared in the RLL (B8B9) and in the

264

remaining 4 lobes in biopsies samples collected in the 15 severe asthmatics 15 days before

265

BT (B). Results represent medians (25-75 interquartile range; minimum and maximum) and

266

horizontal bars show median values. Comparisons in all panels were made using Students’ t-

267

test for paired values.

268

Figure E4. Changes in collagen deposition after BT in severe asthmatics. Exemplary

269

Masson’s trichrome staining in sections from bronchial biopsies of severe asthmatics

270

collected before (A-C) and 3 months after (D-F) BT. A,B,D,E show collagen deposition (in

271

green), and C and F illustrate the corresponding immune-surface algorithm results

272

(blue/negative, yellow/low intensity, orange/moderate intensity and red/high intensity).

273

Original magnifications x 5 (A,D) and x 40 (B,C,E,F). (G) Proportion of biopsy area stained

274

for collagen before and after BT. Data are means ± SEM and comparisons were made by

275

Students’ t-test for paired values. (H) The intensity of collagen deposition was quantified

276

using an immune-surface algorithm that defined four categories of magnitude of expression,

277

i.e., absence = score 0; low = score 1+; moderate = score 2+ and high = score 3+. Open and

278

closed columns, represent results obtained before and after BT, respectively). Data are

279

means ± SEM and comparisons were made using Students’ t-test for paired values.

SC

M AN U

TE D

EP

AC C

280 281

RI PT

261

12

RI PT

ACCEPTED MANUSCRIPT

Table E1. Primary and secondary antibodies for immunohistochemistry Clone*

Supplier

Dilution

Against

CD31

JC70A

Dako

1:100

Blood vessels

D2-40

D2-40

Biocare Medical

1:150

Lymphatic vessels

ECP

EG2

Diagnostic Development

1:1000

Eosinophils

MPO

Rabbit polyclonal

Dako

1:10000

Neutrophils

PGP9.5

10A1

Novocastra

1:60

Sigma-Aldrich

1:1000

Nerves and neuroendocrine cells Smooth muscle actin

SMA

1A4

Secondary HRP-conjugated

Goat polyclonal

Dako

1:200

Mouse Ig

Secondary HRP-conjugated

Goat polyclonal

Invitrogen

1:200

Rabbit IgG

TE D

M AN U

SC

Antibody

*Mouse monoclonal unless otherwise stated.

Abbreviations; ECP, eosinophilic cationic protein; MPO, myeloperoxidase; PGP9.5, protein gene product 9;

AC C

EP

SMA, smooth muscle actin.

RI PT

ACCEPTED MANUSCRIPT

Table E2. Asthma control in severe refractory asthmatics according to their clinical response to BT 3 months after BT Asthma control

Before BT

No. of patients

15

Score on ACT

8.5 ± 2.8

Score on AQLQ

2.6 ± 0.9

Annual rate of severe exacerbations

18.0 ± 3.3

P value

4

a

Responsive

Poorly responsive

11

4

P value

9.5 ± 1.7

< 0.001

19.6 ± 5.0

7.8 ± 2.2

< 0.02

4.2 ± 1.1

2.4 ± 1.7

0.10

4.9 ± 1.0

2.4 ± 0.8

< 0.01

9.7 ± 1.3

0.3 ± 0.2

1.8 ± 0.2

< 0.01

0.3 ± 0.1

2.0 ± 0.7

0.24

1.7 ± 0.8

0

0.8 ± 0.2

0.02

0

0.8 ± 0.2

0.02

3.3 ± 1.0

0.3 ± 0.2

1.3 ± 0.3

0.03

0

0.8 ± 0.4

0.44

0.9 ± 0.7

0

0.8 ± 0.2

0.02

0

0.8 ± 0.3

0.13

4 (36)

4 (100)

10.0 ± 4.1

17.5 ± 2.9

TE D

Annual rate of hospitalization in ICU

b

M AN U

Annual rate of hospitalization for asthma Annual rate of visits to emergency department

11

Poorly responsive

SC

Responsive

12 months after BT

On maintenance use of OCS – no. (%)

10 (67)

5 (45)

4 (100)

Dose of oral prednisone – mg per day

31.5 ± 11.1

17.0 ± 14.0

25.0 ± 10.0

< 0.001 0.25

< 0.001 0.05

AC C

EP

Abbreviations: BT, bronchial thermoplasty; ACT, asthma control test; AQLQ, asthma quality of life questionnaire; ICU, intensive care unit; OCS, oral corticosteroids a One-way ANOVA; * P ≤ 0.05, as compared to responders (Student’s t test for paired values, or Fisher exact test) b Data are n (%), or means ± SD, or means ± SE for annual rates of severe exacerbations, hospitalizations for asthma, visits to emergency department and hospitalizations in ICU c Bold denotes significant difference

a

ACCEPTED MANUSCRIPT

RI PT

Table E3. Expression of hallmarks of airway inflammation and remodeling before and 3 months after BT in untreated and -treated lung lobes

Before BT

After BT (untreated middle lobes)

After BT (treated lobes)

P value

ASM (% of mucosal area)

19.7 (16.2-21.8)

18.0 (6.8-22.5)

5.2 (3.7-9.8)

0.01

SBM thickening (µm)

4.4 (4.0-4.7)

4.4 (4.0-4.7)

3.9 (3.7-4.5)

0.19

1.4 (1.3-2.0)

1.9 (0.7-3.0)

2.3 (1.5-2.7)

0.87

32.5 (21.1-33.8)

34.8 (13.4-45.2)

22.9 (19.0-32.9)

0.40

0.12 (0.07-0.21)

0.12 (0.09-0.36)

0.11 (0.07-0.39)

0.74

1.8 (1.1-2.9)

1.7 (0.8-3.6)

2.0 (1.2-2.7)

1

Sub-mucosal nerves (‰ PGP immunoreactivity)

1.0 (0.8-1.3)

0.5 (0.1-1.8)

0.3 (0.1-0.5)

0.16

ASM-associated PGP+ nerves (pixels per mm2 ASM)

452 (210-755)

318 (48-468)

63 (0-219)

0.01

Glandular tissue (% prevalence b)

20 (0-30)

0

12.5 (0.0-19.5)

0.45

0 (0-16)

0 (0-50)

25.0 (13-36)

0.94

Columnar stratified epithelium (% prevalence)

60 (46-83)

100 (50-100)

67 (44-82)

0.30

Metaplastic epithelium (% prevalence)

22 (17-47)

0

17 (7-34)

0.25

Squamous metaplastic epithelium (% prevalence)

14 (0-25)

0

0 (0-21)

0.77

Goblet cell hyperplasia/hypertrophy (% prevalence)

60 (36-75)

100 (50-100)

60 (50-73)

0.10

Number of PGP+ neuroendocrine cells per mm2 epithelium

4.9 (0.3-14.1)

0

0

1

Sub-mucosal lymphatic vessels (‰ D2-40 immunoreactivity)

Eosinophils (% ECP immunoreactivity) Neutrophils (% MPO immunoreactivity)

EP

TE D

Regenerating epithelium (% prevalence)

M AN U

Sub-mucosal blood vessels (‰ CD31 immunoreactivity)

SC

Parameter

a

AC C

Abbreviations: BT, bronchial thermoplasty; ASM, airway smooth muscle; SBM, sub-epithelial basement membrane; ECP, eosinophil-cationic protein; MPO, myeloperoxidase; PGP, protein gene product (nerve and epithelium neuroendocrine cell marker). a

the differences between pre and post thermoplasty values in both group (treated and not treated) were compared using exact rank sum test.

b

% prevalence (yes/no) in the set of biopsies

ACCEPTED MANUSCRIPT

RI PT

Table E4. Correlation analyses between ASM area and hallmarks of airway inflammation and remodeling before and 3 months after BT Parameter SBM thickening (µm)

P value

0.33

0.22

- 0.41

0.07

SC

Sub-mucosal lymphatic vessels (‰ D2-40 immunoreactivity)

r

Sub-mucosal blood vessels (‰ CD31 immunoreactivity)

0.07

0.91

Eosinophils (% ECP immunoreactivity)

0.17

0.76

- 0.08

0.98

M AN U

Neutrophils (% MPO immunoreactivity)

0.56

0.007 b

ASM-associated PGP+ nerves (pixels per mm2 ASM)

0.44

0.07

Glandular tissue (% prevalence c)

0.28

0.33

Regenerating epithelium (% prevalence)

- 0.10

0.87

Columnar stratified epithelium (% prevalence)

- 0.00

0.98

Metaplastic epithelium (% prevalence)

- 0.14

0.82

Squamous metaplastic epithelium (% prevalence)

- 0.01

0.87

Goblet cell hyperplasia/hypertrophy (% prevalence)

0.05

0.91

Number of PGP+ neuroendocrine cells per mm2 epithelium

0.75

0.007

EP

TE D

Sub-mucosal nerves (‰ PGP immunoreactivity)

a

AC C

Abbreviations: ASM, airway smooth muscle; BT, bronchial thermoplasty SBM, sub-epithelial basement membrane; ECP, eosinophil-cationic protein; MPO, myeloperoxidase PGP, protein gene product (nerve and epithelium neuroendocrine cell marker); a Spearman’s rank-order method and the Benjamini and Hochberg correction b Bold denotes significant correlation c % prevalence (yes/no) in the set of biopsies

ACCEPTED MANUSCRIPT

RI PT

Table E5. Correlation analyses between histopathological markers and the cumulative numbers of heat-activations during BT Parameter

r

- 0.48

0.83

- 0.28

0.88

Sub-mucosal blood vessels (‰ CD31 immunoreactivity)

0.26

0.88

Eosinophils (% ECP immunoreactivity)

0.02

0.99

Neutrophils (% MPO immunoreactivity)

- 0.12

0.99

SC

SBM thickening (µm)

P value

M AN U

Sub-mucosal lymphatic vessels (‰ D2-40 immunoreactivity)

Sub-mucosal nerves (‰ PGP immunoreactivity)

0.001

0.99

ASM-associated PGP+ nerves (pixels per mm2 ASM)

0.06

0.99

- 0.07

0.99

0.21

0.94

0.01

0.99

- 0.43

0.83

Squamous metaplastic epithelium (% prevalence)

0.03

0.99

Goblet cell hyperplasia/hypertrophy (% prevalence)

0.35

0.86

Number of PGP+ neuroendocrine cells per mm2 epithelium

- 0.33

0.86

SBM thickening (µm)

- 0.17

0.99

Glandular tissue (% prevalence b)

TE D

Regenerating epithelium (% prevalence)

Columnar stratified epithelium (% prevalence)

AC C

EP

Metaplastic epithelium (% prevalence)

a

Abbreviations: BT, bronchial thermoplasty; ASM, airway smooth muscle; SBM, sub-epithelial basement membrane; ECP, eosinophil-cationic protein; MPO, myeloperoxidase; PGP, protein gene product (nerve and epithelium neuroendocrine cell marker). a Spearman’s rank-order method and the Benjamini and Hochberg correction b % prevalence (yes/no) in the set of biopsies

ACCEPTED MANUSCRIPT

RI PT

Table E6. Correlation analyses between airway smooth muscle area and clinical parameters 3 months after BT

Parameter

r

P value

0.14

0.99

- 0.04

0.99

- 0.00

0.99

0.07

0.99

0.03

0.99

0.06

0.99

- 0.03

0.99

0.01

0.99

Pre-bronchodilator FEV1/FVC (percentage of predicted)

0.10

0.99

Post-bronchodilator FEV1/FVC (percentage of predicted)

0.08

0.99

Reversibility to β2-agonists (mL)

0.18

0.99

Reversibility to β2-agonists (%)

0.17

0.99

9

Blood eosinophils (10 /L)

SC

Total serum IgE (IU/mL) Pre-bronchodilator FEV1 (mL) Pre-bronchodilator FEV1 (% predicted) Post-bronchodilator FEV1 (% predicted) Pre-bronchodilator FVC (mL)

TE D

Post-bronchodilator FVC (mL)

M AN U

Post-bronchodilator FEV1 (mL)

a

AC C

EP

Abbreviations: BT, bronchial thermoplasty; FEV1, Forced Expiratory Volume in One second; FVC, Forced Vital Capacity a Spearman’s rank-order method and the Benjamini and Hochberg correction b Bold denotes statistical significance

ACCEPTED MANUSCRIPT

Table E7. Correlation analyses between SBM thickening and clinical parameters 3 months after BT

r

RI PT

Parameter 9

Blood eosinophils (10 /L) Total serum IgE (IU/mL) Pre-bronchodilator FEV1 (mL)

- 0.09

0.82

0.22

0.75

- 0.15

0.82

- 0.06

0.82

SC

Post-bronchodilator FEV1 (mL) Pre-bronchodilator FEV1 (% predicted)

P value

0.02

0.93

0.08

0.82

- 0.22

0.75

- 0.16

0.82

Pre-bronchodilator FEV1/FVC (percentage of predicted)

0.13

0.82

Post-bronchodilator FEV1/FVC (percentage of predicted)

0.09

0.82

Reversibility to β2-agonists (mL)

0.22

0.75

0.24

0.75

Pre-bronchodilator FVC (mL) Post-bronchodilator FVC (mL)

TE D

Reversibility to β2-agonists (%)

M AN U

Post-bronchodilator FEV1 (% predicted)

a

AC C

EP

Abbreviations: SBM, sub-epithelial basement membrane; BT, bronchial thermoplasty; FEV1, Forced Expiratory Volume in One second; FVC, Forced Vital Capacity a Spearman’s rank-order method and the Benjamini and Hochberg correction

ACCEPTED MANUSCRIPT

RI PT

Table E8. Correlation analyses between sub-mucosal nerves (PGP+ cells) and clinical parameters 3 months after BT

Parameter

r

9

P value

- 0.124

0.97

Total serum IgE (IU/mL)

- 0.053

0.97

SC

Blood eosinophils (10 /L) Pre-bronchodilator FEV1 (mL) Pre-bronchodilator FEV1 (% predicted)

0.97

- 0.068

0.97

M AN U

Post-bronchodilator FEV1 (mL)

- 0.051 0.010

0.97

- 0.092

0.97

0.008

0.97

0.053

0.97

Pre-bronchodilator FEV1/FVC (percentage of predicted)

0.042

0.97

Post-bronchodilator FEV1/FVC (percentage of predicted)

- 0.032

0.97

Reversibility to β2-agonists (mL)

0.102

0.97

Reversibility to β2-agonists (%)

0.106

0.97

Post-bronchodilator FEV1 (% predicted) Pre-bronchodilator FVC (mL)

TE D

Post-bronchodilator FVC (mL)

a

AC C

EP

Abbreviations: BT, bronchial thermoplasty; PGP, protein gene product; FEV1, Forced Expiratory Volume in One second; FVC, Forced Vital Capacity a Spearman’s rank-order method and the Benjamini and Hochberg correction

ACCEPTED MANUSCRIPT

RI PT

Table E9. Correlation analyses between PGP+-associated airway smooth muscle nerves and clinical parameters 3 months after BT

Parameter

r

P value

0.126

0.98

- 0.219

0.90

- 0.210

0.90

- 0.198

0.90

- 0.028

0.98

- 0.199

0.90

- 0.052

0.98

- 0.050

0.98

- 0.024

0.98

- 0.065

0.98

Reversibility to β2-agonists (mL)

0.072

0.98

Reversibility to β2-agonists (%)

0.055

1

9

Blood eosinophils (10 /L)

SC

Total serum IgE (IU/mL) Post-bronchodilator FEV1 (mL) Pre-bronchodilator FEV1 (% predicted) Post-bronchodilator FEV1 (% predicted) Pre-bronchodilator FVC (mL) Post-bronchodilator FVC (mL)

M AN U

Pre-bronchodilator FEV1 (mL)

TE D

Pre-bronchodilator FEV1/FVC (percentage of predicted)

EP

Post-bronchodilator FEV1/FVC (percentage of predicted)

a

AC C

Abbreviations: BT, bronchial thermoplasty; PGP, protein gene product (nerve and epithelium neuroendocrine cell marker); FEV1, Forced Expiratory Volume in One second; FVC, Forced Vital Capacity a Spearman’s rank-order method and the Benjamini and Hochberg correction

ACCEPTED MANUSCRIPT

RI PT

Table E10. Correlation analyses between PGP+ epithelium neuroendocrine cells and clinical parameters 3 months after BT

Parameter

r

P value

0.036

0.86

- 0.274

0.56

- 0.201

0.86

- 0.077

0.86

- 0.100

0.86

- 0.118

0.86

- 0.070

0.86

- 0.034

0.86

- 0.058

0.86

- 0.067

0.86

Reversibility to β2-agonists (mL)

0.309

0.56

Reversibility to β2-agonists (%)

0.305

0.56

9

Blood eosinophils (10 /L)

SC

Total serum IgE (IU/mL) Post-bronchodilator FEV1 (mL) Pre-bronchodilator FEV1 (% predicted) Post-bronchodilator FEV1 (% predicted) Pre-bronchodilator FVC (mL) Post-bronchodilator FVC (mL)

M AN U

Pre-bronchodilator FEV1 (mL)

TE D

Pre-bronchodilator FEV1/FVC (percentage of predicted)

EP

Post-bronchodilator FEV1/FVC (percentage of predicted)

a

AC C

Abbreviations: BT, bronchial thermoplasty; PGP, protein gene product (nerve and epithelium neuroendocrine cell marker); FEV1, Forced Expiratory Volume in One second; FVC, Forced Vital Capacity a Spearman’s rank-order method and the Benjamini and Hochberg correction b Bold denotes significant correlations

ACCEPTED MANUSCRIPT

Table E11: Stability of clinical characteristics in severe therapy-uncontrolled severe asthmatics over time Values at 4th visit of follow-up

20 14 (70) 42.1 ± 13.0 17 (75) 10 (50) 26.3 ± 4.0 17.3 ± 14.6 24.8 ± 18.2 7 (35) 148 (100-284) 353 (64-787)

20

63.7 ± 21.3 69.8 ± 21.1 1.96 ± 0.75 2.14 ± 0.76 60.8 ± 12.9 6.1 ± 6.5 188 ± 214

64.9 ± 21.2 71.4 ± 20.2 1.97 ± 0.73 2.13 ± 0.71 61.0 ± 12.1 6.4 ± 4.6 159 ± 127

20 (100) 2905 ± 1074 5 (25) 22.0 ± 21.7

20 (100) 2747 ± 1123 8 (40)

AC C

EP

TE D

P valueb

___

RI PT

M AN U

Number of patients Female sex – no. (%) Age (years) Caucasian origin – no. (%) Former smokers – no. (%) 2 Body Mass Index (kg per m ) Age of asthma onset (years) Asthma duration (years) History of atopy – no. (%) 9 Blood eosinophils (10 /L) Total serum IgE (International Units/mL) Respiratory function Pre-bronchodilator FEV1 (% predicted) Post-bronchodilator FEV1 (% predicted) Pre-bronchodilator FEV1 (L) Post-bronchodilator FEV1 (L) Pre-bronchodilator FEV1 / FVC (% predicted) Reversibility to β2-agonists (%) Reversibility to β2-agonists (mL) Treatments On long-acting β2-agonists – no. (%) Dose of inhaled corticosteroids (µg equivalents of beclomethasone per day) On maintenance use of oral corticosteroids (%) Daily dose of oral corticosteroids in mg On anti-IgE – no. (%) Asthma control Score on asthma control test Annual rate of severe exacerbations

Values at 2nd visit a of follow-up

SC

Parameter

___ ___ ___ ___ ___ ___ ___ ___ ___

9 (45)

29.4 ± 21.1 9 (45)

9.6 ± 3.2 4.1 ± 0.7

9.5 ± 2.8 4.0 ± 0.4

Abbreviations: FEV1, Forced Expiratory Volume in One second; FVC, Forced Vital Capacity a Data are n (%), or means ± SD, or medians (25-75 interquartile range), or means ± SE for annual rate of severe exacerbations b Student’s t test for unpaired values, of Fisher exact test (two-tailed)

0.86 0.82 0.95 0.96 0.97 0.86 0.61

0.66 0.03 0.56

0.91 0.94

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure E1

ACCEPTED MANUSCRIPT

B

P = 0.103

P = 0.991

0.15

M AN U

Epithelium area (mm2)

1.5

1.0

0.0 Before BT

EP

TE D

0.5

After BT

AC C

Total biopsy area (mm2)

SC

A

RI PT

Figure E2

0.10

0.05

0.00 Before BT

After BT

ACCEPTED MANUSCRIPT

Figure E3 B

40

20

30 20 10

0

Before BT

AC C

EP

P = 0.53

TE D

Before entry

D

Before BT RLL (B8B9)

Before BT (remaining 9 lobes)

P = 0.35

8000 ASM PGP (pixels/mm2)

0

C

40

RI PT

ASM area (% mucosa area)

60

P = 0.12

50

M AN U

ASM area (% mucosa area)

P = 0.30

SC

A

6000 4000 2000 0

Before entry

Before BT

Before entry

Before BT

Collagen deposition (% of total biopsy area) 80

60

40

20

0

Score 0

P = 0.002 50

40

30

20

10

0

RI PT

Collagen deposition (% of total biopsy area)

SC

F

M AN U

C 100

Score 1+

P = 0.003

50

40

30

20

10

0

80

Collagen deposition (% of total biopsy area)

H

TE D

E

Collagen deposition (% of total biopsy area)

100

B

EP

D

AC C

A

Collagen deposition (% of total biopsy area)

ACCEPTED MANUSCRIPT

Figure E4 G Before BT

P = 0.005

After BT

P = 0.002

60

40

20 0

Score 2+ Score 3+

50

40

30

20

10 0

P = 0.009