In vivo confocal microscopy of dermoscopic suspicious lesions in patients with xeroderma pigmentosum: A cross-sectional study

Journal Pre-proof In vivo confocal microscopy of dermoscopic suspicious lesions in Xeroderma Pigmentosum Patients – A Cross-sectional Study Lílian Kelly Faria Licarião Rocha, MD, Paula Ferreira, MD, João Avancini, MD, Silvia V. Lourenço, PhD, Caroline F. Barbosa, MD, Caroline Colacique, MD, Cyro Festa Neto, PhD PII:

S0190-9622(19)33293-1

DOI:

https://doi.org/10.1016/j.jaad.2019.12.018

Reference:

YMJD 14069

To appear in:

Journal of the American Academy of Dermatology

Received Date: 2 July 2019 Revised Date:

1 November 2019

Accepted Date: 11 December 2019

Please cite this article as: Rocha LKFL, Ferreira P, Avancini J, Lourenço SV, Barbosa CF, Colacique C, Neto CF, In vivo confocal microscopy of dermoscopic suspicious lesions in Xeroderma Pigmentosum Patients – A Cross-sectional Study, Journal of the American Academy of Dermatology (2020), doi: https://doi.org/10.1016/j.jaad.2019.12.018. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier on behalf of the American Academy of Dermatology, Inc.

1

TITLE: In vivo confocal microscopy of dermoscopic suspicious

2

lesions in Xeroderma Pigmentosum Patients – A Cross-sectional

3

Study

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

AUTHORS: Rocha, L.; Ferreira, P.; Avancini J.; Lourenço V. S.; Barbosa F. C.; Colacique C.; Festa-Neto C. Dermatology Department of Hospital das Clínicas of São Paulo University, Brazil

-Lílian Kelly Faria Licarião Rocha: MD / Researcher and Dermatologist at Department of Dermatology, Hospital das Clínicas of São Paulo University, São Paulo, Brazil. -Paula Ferreira: MD / Dermatologist at Hospital das Clínicas of São Paulo University, São Paulo, Brazil. -João Avancini: MD / Dermatologist at Hospital das Clínicas of São Paulo University, São Paulo, Brazil. -Silvia V Lourenço: PhD / Dentist at Hospital das Clínicas of São Paulo University, São Paulo, Brazil. -Caroline F Barbosa: MD / Dermatologist at Hospital das Clínicas of São Paulo University, São Paulo, Brazil. -Caroline Colacique: MD / Dermatologist at Hospital das Clínicas of São Paulo University, São Paulo, Brazil. -Cyro Festa Neto: PhD/ Professor at Department of Dermatology, Hospital das Clínicas of São Paulo University, São Paulo, Brazil.

CORRESPONDING AUTHOR:

Lílian Rocha E-mail: [email protected] Address: Av. Dr. Enéas de Carvalho, 255 São Paulo - Brazil

WORD COUNT: Capsule Summary: 42 Abstract: 199 Article: 1960 NUMBER OF FIGURES: 1 (1a, 1b, 1c, 1d) NUMBER OF TABLES: 2 NUMBER OF REFERENCES: 17 Financial disclosure: none reported Funding/support: this study had no funding and financial support. The authors have no conflict of interest to disclose.

47 48 49

ABSTRACT

50 51

INTRODUCTION: Xeroderma pigmentosum (XP) is a rare genetic disease that is

52

characterized by extreme photosensitivity, resulting in a higher incidence of cutaneous

53

tumors. Reflectance confocal microscopy (RCM) is a noninvasive imaging method for

54

diagnosing cutaneous lesions.

55

OBJECTIVE: To explore the application of RCM in the follow-up of XP patients.

56

MATERIALS AND METHODS: XP patients underwent RCM for suspicious lesions from

57

January 2010 to April 2019. Lesions with malignant RCM features were excised, and the

58

results were compared with their histopathological features. Benign lesions on RCM were

59

monitored every 3 months. We recorded the confocal features that were related to

60

malignancy and specifically to melanoma.

61

RESULTS: A total of 61 suspicious lesions from 13 XP patients were included. Thirty-three

62

lesions (54%) were malignant (14 melanomas, 15 basal cell carcinomas, and 4 squamous cell

63

carcinomas).

64

Nonvisible papillae (OR: 11.8 [95% CI 2.6-53.1], p:0.001) and atypical cells at the dermal-

65

epidermal junction (OR: 11.7 [95% CI 2.7-50.3], p:0.001) were independent predictors of

66

malignancy.

67

LIMITATIONS: The limited number of patients and lesions. Most cases were

68

retrospectively included, and some did not undergo a histological analysis.

69

CONCLUSIONS: RCM is a valuable tool in the follow-up of XP patients, reducing the need

70

for excisions by 35%.

71 72

73 74 75

ABBREVIATIONS AND ACRONYMS:

76

BCC: Basal cell carcinoma

77

DEJ: Dermal-epidermal junction

78

OR: Oddis ratio

79

RCM: Reflectance confocal microscopy

80

RCM melanoma score: Confocal microscopic score for melanoma

81

RCM LM score: Confocal microscopic score for lentigo maligna

82

SCC: Squamous cell carcinoma

83

TBP: Total-body photographs

84

XP: Xeroderma pigmentosum

85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105

106

INTRODUCTION:

107

Xeroderma pigmentosum (XP) is a rare disorder in which UV radiation induced

108

damage repair is defective and is characterized by photosensitivity, with a significant

109

predilection for skin burning following minimal sun exposure, abnormal lentiginosis (freckle-

110

like pigmentation) on sun-exposed areas and resulting areas of increased or decreased

111

pigmentation, skin aging, and multiple skin cancers.1 XP patients are estimated to have a

112

10,000-fold greater risk of nonmelanoma skin cancer and a 2,000-fold higher risk for

113

melanoma under the age of 20 years compared with the general population.2

114

The prognosis of XP patients depends on UV photoprotection, a strict clinical follow-

115

up, and the early diagnosis and surgical treatment of skin tumors. Unfortunately, the clinical

116

diagnosis of skin tumors at an early stage and the discrimination between malignant and

117

benign lesions are difficult due to the presence of severe UV-induced skin damage and

118

general actinic lesions.

119

In vivo reflectance confocal microscopy (RCM) is a novel technique that can be used

120

to examine the epidermis and papillary dermis noninvasively with cellular resolution3.

121

Dermoscopy plus RCM is significantly more sensitive and specific than dermoscopy alone in

122

the diagnosis of melanoma and nonmelanoma tumors4, 5, 6,7.

123

This study examined whether RCM in the follow-up of XP patients improves the

124

accuracy of the diagnosis of malignant lesions and reduces the number of unnecessary

125

excisions in this challenging population.

126 127

METHODS

128

This cross-sectional study was performed from January 2010 to April 2019. XP

129

patients who were followed by the tertiary center in the Dermatology Department of Hospital

130

das Clínicas, São Paulo University (Brazil) were enrolled. The study was approved by the

131

national and local ethics committees (CEP - CAAE: 98968918.0.0000.0068, parecer:

132

3.163.255).

133

XP patients were followed every 3 months with a complete cutaneous surface

134

examination, digital dermoscopy, and total-body photographs (TBPs) as standard care. We

135

added an evaluation by RCM for suspicious lesions in XP patients.

136

Dermoscopic images and TBPs were captured using the bodystudio ATBM® method

137

(Fotofinder system, Germany). RCM images were acquired primarily by a Vivascope 3000

138

(Caliber ID, Rochester, NY), which uses an 830-nm laser with a maximum power of 20 mW.

139

Patient demographics and tumor characteristics were recorded, and 2 independent

140

investigators (L.R. and P.F.) evaluated all clinical, dermoscopic, and RCM images. All

141

lesions with at least 1 of the following criteria were considered to be suspicious: any

142

dermoscopic feature that was related to malignancy per previous dermoscopic criteria for

143

melanocytic lesions, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC) 8, 9, 10,

144

11

; any significant change on the TBP; and new lesions on the TBP.

145

RCM analysis was based on microscopic features for melanoma per Pellacani et al.4

146

and Guitera et al.5, and the scores for melanoma (RCM melanoma score4) and lentigo

147

maligna (RCM LM score5) were calculated for each lesion. RCM features for BCC per

148

Guitera P et al.12 and Longo C et al.13 and RCM features for SCC per Manfredini M et al.14

149

and Shahriari N et al.15 were included in this study.

150

All suspicious lesions were evaluated by RCM, and excisions were performed for

151

lesions that were suspicious by RCM. Histological results were compared with the evaluation

152

by RCM. Lesions classified as benign on RCM were not excised and they were monitored

153

every 3 months by dermoscopy and RCM.

154 155

Statistical analysis

156

Data were analyzed using SPSS (version 25.0.0, Windows©). Fisher’s exact and Pearson chi-

157

square tests were used to analyze the association between quantitative factors. Descriptive

158

statistics were calculated for clinical, dermoscopic, and confocal features.

159

Three scenarios were considered in the statistical analysis: “malignant lesions,” which

160

included melanomas, BCCs, and SCCs, “melanoma diagnosis,” which comprised only

161

melanocytic malignant lesions, and “BCC diagnosis”. Logistic regression models were used

162

to test the associations of confocal features with malignancy, the diagnosis of melanoma and

163

BCC diagnosis. Odds ratios (ORs) with p <0.05 were considered to be statistically

164

significant.

165 166

RESULTS

167

A total of 61 suspicious lesions from 13 XP patients were included. Nine patients

168

(69%) were female, and the median age was 27.3 years (range: 9-69 years), with half of the

169

patients aged under 20. The average number of lesions per patient was 5 (range: 1-11

170

lesions), and the main areas that were affected were the face and neck (54%), trunk (24%),

171

and lips and oral area (13%).

172

Of the 61 lesions, 32 were included due to having suspicious dermoscopic features on

173

initial evaluation, 25 presented changes during the dermoscopic monitoring, and 4 were new

174

lesions on the TBP.

175

Thirty-three lesions (54%) were diagnosed as being malignant (14 melanomas, 15

176

BCC, and 4 SCC). Of the 14 melanomas, 13 lesions were in situ and 1 melanoma had a

177

Breslow thickness of 0.2 mm.

178

Twenty-eight lesions were classified as benign (with no previously described feature

179

of malignancy by RCM4, 5, 12, 13, 14, 15): 20 lesions were not excised and were then followed

180

with dermoscopy, TBP, and RCM every 3 months (median follow-up time: 17 months, range

181

5-324 months) but did not present with any significant change. The other 8 benign lesions

182

were excised: 7 were suspicious by RCM (the histological diagnoses were: 5 lesions were

183

atypical intradermal melanocytic hyperplasia, 1 was a junctional melanocytic nevus that was

184

associated with actinic keratosis, and 1 was a dysplastic junctional nevus with pigment

185

incontinence), and the remaining benign lesion was excised as requested by the patient (the

186

histological diagnosis was seborrheic keratosis).

187

The confocal examination revealed features that were related to cutaneous

188

malignancy: epidermal disarray (OR: 10.0 [95% CI: 2.0-50.6], p:0.005), nonvisible papillae

189

(OR: 12.0 [95% CI: 3.3-43.3], p<0.001), atypical cells at the dermal-epidermal junction

190

(DEJ) (OR: 11.8 [95% CI: 3.4-40.8], p<0.001), distribution of bright cells at the DEJ (OR:

191

5.0 [95% CI: 1.7-14.9], p:0.004), dermal involvement (OR: 25.4 [95% CI: 3.1-209.5],

192

p:0.003), presence of pleomorphic cells (OR: 9.4 [95% CI: 2.9-30.2], p<0.001), and linear-

193

horizontal vessels (OR: 8.6 [95% CI: 1.0-74.1], p=0.049). The presence of melanophages

194

(OR: 0.2 [95% CI: 0.1-0.6], p=0.005) and the “cord-like sign”—polycyclic papillae, as

195

circumvolution of edged papillae (OR: 0.2 [95%CI: 0.1-0.8], p=0.019)—were inversely

196

associated with cutaneous malignancy (TABLE I).

197

The RCM features that were statistically associated with a melanoma diagnosis were:

198

pagetoid cells (OR: 5.9 [95%CI: 1.1-30.5], p=0.035), adnexal spread (OR: 9.0 [95%CI: 1.4-

199

56.1], p=0.019), atypical cells at the DEJ (OR: 5.3 [95%CI: 1.1-26.2], p=0.042), nests of

200

melanocytes (OR: 18.4 [95%CI: 1.9-182.7], p=0.013), irregular distribution of bright cells

201

(OR: 9.7 [95%CI: 1.9-48.3], p=0.006), and irregular distribution of the pigmentation (OR:

202

5.3 [95%CI: 1.1-26.2], p=0.042). The sensitivity and specificity of RCM melanoma score4

203

were 85.7% and 46.8% for a melanoma diagnosis, compared with 57.1% and 76.6% of RCM

204

LM score5, respectively. TABLE II

205

Confocal analysis of BCC lesions showed that most previously described RCM

206

features for BCC15,16 were present in these lesions and were statistically associated with a

207

BCC diagnosis: plump-bright cells (OR: 60.5 [95%CI: 9.8-373.9], p<0.001), streaming

208

epidermis (OR: 33.0 [95%CI: 5.7-190.6], p<0.001), dark silhouette and clefting (OR: 44.0

209

[95%CI: 7.4-260.3], p<0.001), linear telangiectasia-like horizontal vessel (OR: 19.3

210

[95%CI:3.4-110.0], p=0.001), and peripheral palisading (OR:11.0 [95%CI: 1.9-65.1],

211

p=0.08). Of the 15 BCC lesions, 40% (6/15) were reported histologically as pigmented. By

212

dermoscopy, most of the BCC lesions were pigmented: 86.7% (13/15) presented with

213

irregular globules and dots, 73.3% (11/15) presented with an asymmetrical blotch, and 60%

214

(9/15) were grey.

215

Some cases of melanoma showed RCM features that resembling BCC: 2 cases

216

(14,3%) had streaming epidermis, plump-bright cells, peripheral palisading, dark silhouette,

217

and clefting, and 1 case (7%) showed a linear telangiectasia-like horizontal vessel.

218

Among the 4 SCC cases, 2 lesions were histological in situ and 2 were superficial

219

invasive. All SCC cases presented with epidermal disarray and looping vessels, 3 cases

220

presented with round circles with a bright white rim and inflammation, and 2 cases harbored

221

erosion and dendritic cells.

222

All 61 lesions in this study were suspicious lesions on dermoscopy. They would have

223

been removed, regardless of intervention by RCM. Only lesions suspicious on RCM (total of

224

40/61) were removed and had histological analysis. The remaining 21 lesions were not

225

excised and were monitored by dermoscopy and RCM. During monitoring these lesions

226

presented no changes and they were classified as benign. Ultimately, 65% (40/61) of the

227

lesions were classified as being malignant by RCM and were excised, decreasing the number

228

of excisions that was needed by 35%.

229 230

Multivariate analysis

231

The multivariate logistic regression model showed that by RCM, nonvisible papillae (OR:

232

11.8 [95% CI 2.6-53.1], p:0.001) and atypical cells at the DEJ (OR: 11.7 [95% CI 2.7-50.3],

233

p:0.001) were independent predictors of malignancy.

234

235 236 237

DISCUSSION

238

We have reported confocal features that are related to suspicious lesions in XP

239

patients. Most of our patients were female, and half of them were aged under 20 years. The

240

early onset of skin cancer is one of the main characteristics of XP, and the average age at the

241

first diagnosis of a cutaneous tumor is 8 years1. As expected, our sample presented with

242

severe sun-induced damage with multiple hyperpigmented, hypopigmented, and scaly

243

lesions, making the clinical diagnosis of cutaneous malignancy difficult.

244

Dermoscopy, TBP, and RCM increase the diagnostic accuracy of malignancies,

245

particularly melanoma3, 4, 5, 6, 7, 16. We suggest that the combination of these noninvasive tools

246

will improve the accuracy of such diagnoses in XP patients and should be implemented in the

247

follow-up of XP patients.

248

Harris Green et al.17 reported a case with XP who was followed monthly by clinical

249

and dermoscopic examination in association with TBPs. They reported that the thickest

250

melanoma in their series was 1.07 mm and that the mean Breslow thickness was 0.18mm.

251

Most melanomas in our series were in situ, and only 1 melanoma had a Breslow thickness of

252

0.2 mm. We suggest that our approach of combining total-body cutaneous examination, TBP,

253

and dermoscopy every 3 months, in association with RCM examination, for every suspicious

254

lesion will improve the early detection of melanoma in XP patients.

255

The most important malignant differential diagnosis of melanoma was pigmented

256

BCC in our series. Most BCC lesions presented with irregular globules and dots,

257

asymmetrical blotches, and grey color, complicating the clinical differentiation of these 2

258

diagnoses. RCM was a good approach for differentiating melanoma from pigmented BCC in

259

this study. The RCM features that have been described for BCC15,16 were seen primarily in

260

BCC tumors. Further, pleomorphic cells and linear-horizontal vessels were predictors of

261

malignancy but not melanoma.

262

The sensitivity of the RCM melanoma score4 in this series was lower compared with

263

Pellacani et al.4 (85.7% and 96.3%, respectively), as was the specificity (46.8% vs 52.1%) for

264

a melanoma diagnosis. However, Pellacani et al. 4 described the score only for melanocytic

265

lesions in the general population.

266

Our approach in XP patients led to a correct diagnosis of malignancy in 100% of

267

patients (all 33 cases of malignancy were diagnosed correctly by RCM), with a positive

268

predictive value of 82.5% (7 lesions that were classified as malignant by RCM were benign

269

by histology), and the number of excisions that was reduced by 35%. Notably, the false-

270

positive cases (lesions that were classified as malignant by RCM but nonmalignant by

271

histology) in our series were lesions with cytological atypia.

272

Our study had several limitations. The sample of XP patients was small, and the

273

number of malignant lesions was also limited. Most benign lesions did not undergo a

274

histological evaluation and were followed using noninvasive techniques. Most cases were

275

evaluated retrospectively.

276

There has been no previous report on the use of RCM in clinical decision-making for

277

XP patients. We found that RCM, in association with total-body cutaneous examination,

278

TBP, and dermoscopy, in XP patients improves the accuracy of the diagnosis of malignancy

279

and can help differentiate the main cutaneous diagnosis, leading to a lower rate of excision

280

and perhaps the early detection of melanoma. Future research should determine whether TBP

281

and dermoscopy every 3 months, combined with RCM, accelerates the early detection of

282

melanoma in XP patients.

283 284 285

286 287 288

REFERENCES:

289 290

1. Alan R Lehmann, David McGibbon and Miria Stefanini. Xeroderma pigmentosum.

291

Orphanet Journal of Rare Diseases 2011, 6:70.

292 293

2. Bradford PT, Goldstein AM, Tamura D, et al. Cancer and neurologic degeneration in xeroderma

294

pigmentosum: long term follow-up characterises the role of DNA repair. J Med Genet 2011, 48:168-

295

176.

296 297

3. Rajadhyaksha M, Grossman M, Esterowitz D, et al. In vivo confocal scanning laser microscopy of

298

human skin: melanin provides strong contrast. J Invest Dermatol. 1995 Jun;104(6):946-52.

299 300

4. Giovanni Pellacani, Pascale Guitera, Caterina Longo et al. The Impact of In Vivo Reflectance

301

Confocal Microscopy for the Diagnostic Accuracy of Melanoma and Equivocal Melanocytic Lesions.

302

Journal of Investigative Dermatology (2007) 127, 2759–2765.

303 304

5. Guitera P, Pellacani G, Crotty KA, et al. The impact of in vivo reflectance confocal microscopy on

305

the diagnostic accuracy of lentigo maligna and equivocal pigmented and nonpigmented macules of

306

the face. J Invest Dermatol. 2010 Aug;130(8):2080-91. doi: 10.1038/jid.2010.84. Epub 2010 Apr 15.

307 308

6. Rao BK, Mateus R, Wassef C et al. In vivo confocal microscopy in clinical practice: comparison

309

of bedside diagnostic accuracy of a trained physician and distant diagnosis of an expert reader. J

310

Am Acad Dermatol. 2013 Dec;69(6):e295-300. doi: 10.1016/j.jaad.2013.07.022. Epub 2013 Sep 13.

311

312

7. Alarcon I, Carrera C, Palou J et al. Impact of in vivo reflectance confocal microscopy on the

313

number needed to treat melanoma in doubtful lesions. Br J Dermatol. 2014 Apr;170(4):802-8. doi:

314

10.1111/bjd.12678.

315 316

8. Argenziano G, Soyer HP, Chimenti S et al. Dermoscopy of pigmented skin lesions. Results of a

317

consensus meeting via the internet. J Am Acad Dermatol 2003; 48: 679-93.

318 319

9: Menzies SW, Westerhoff K, Rabinovitz H et al. Surface microscopy of pigmented basal cell

320

carcinoma. Arch Dermatol 2000; 136: 1012–16.

321 322

10: Zalaudek I, Giacomel J, Schmid K et al. Dermatoscopy of facial actinic keratosis, intraepidermal

323

carcinoma, and invasive squamous cell carcinoma: A progression model. J Am Acad Dermatol 2012;

324

Volume 66, Issue 4, Pages 589–597

325 326

11. Lallas A, Tschandl P, Kyrgidis A et al. Dermoscopic clues to differentiate facial lentigo maligna

327

from pigmented actinic keratosis. Br J Dermatol. 2016 May;174(5):1079-85. doi: 10.1111/bjd.14355.

328 329

12. Guitera P, Menzies S, Longo C et al. In Vivo Confocal Microscopy for Diagnosis of Melanoma

330

and Basal Cell Carcinoma Using a Two-Step Method: Analysis of 710 Consecutive Clinically

331

Equivocal Cases. Journal of Investigative Dermatology (2012), Volume 132

332 333

13. Longo C, Lallas A, Kyrgidis A et al. Classifying distinct basal cell carcinoma subtype by means

334

of dermatoscopy and reflectance confocal microscopy. J Am Acad Dermatol, October 2014; Volume

335

71, Issue 4, Pages 716–724.

336

337

14. Manfredini M, Longo C, Ferrari B et al. Dermoscopic and reflectance confocal microscopy

338

features of cutaneous squamous cell carcinoma. J Eur Acad Dermatol Venereol. 2017

339

Nov;31(11):1828-1833.

340 341

15. Shahriari N, Grant-Kels JM, Rabinovitz HS et al. Reflectance Confocal Microscopy Criteria of

342

Pigmented Squamous Cell Carcinoma In Situ. Am J Dermatopathol. 2018 Mar;40(3):173-179.

343 344

16. Wang SQ, Kopf AW, Koenig K. Detection of melanomas in patients followed up with total

345

cutaneous examinations, total cutaneous photography, and dermoscopy. J Am Acad Dermatol. 2004

346

Jan;50(1):15-20.

347 348

17. W. Harris Green; Steven Q. Wang; Armand B. Cognetta Jr et al. Total-Body Cutaneous

349

Examination, Total-Body Photography, and Dermoscopy in the Care of a Patient with Xeroderma

350

Pigmentosum and Multiple Melanomas. Arch Dermatol. 2009 Aug;145(8):910-5.

351 352 353 354

LEGEND

355

Fig 1. Xeroderma Pigmentosum. Example of lesions in Xeroderma Pigmentosum patients

356

(A) Dermoscopy of a benign lesion showing atypical network and asymmetrical globules and

357

dots (B) Confocal microscopy of the benign lesion with pigmented small bright cells and no

358

atypical cells at DEJ (C) Dermoscopy of a melanoma showing asymmetric hyperpigmented

359

follicular opening, annular-granular pattern, dust-like image and increased density of the

360

vascular network (D) Confocal microscopy of the melanoma with epidermal disarray and

361

irregular diamond-shaped large cells (>20µ) in the epidermis (pagetoid cells).

362

363 364 365 TABLE I: Confocal features related to cutaneous malignancy in Xeroderma Pigmentosum patients. Characteristics

Univariate logistic regression

p-value

OR (CI95%)

Epidermal disarray

Present

10.0 (2.0-50.6)

0.005

Irregular diamond-shaped large cells (>20µ µ) in the epidermis (pagetoid cells)

Present

6.0 (0.7-53.2)

0.108

Non-edged papillae

Present

2.6 (0.8-8.7)

0.114

Non-visible papillae

Present

12.0 (3.3-43.3)

<0.001

Atypical cells at DEJ

Present

11.8 (3.4-40.8)

<0.001

Round cells with dark nucleus at the junction size

Small (= the size of epidermal cells around)

0.2 (0.1-0.99)

0.049

Large (twice the size of epidermal cells around)

1.3 (0.3-5.5)

0.676

Mild (<5 pleomorphic2.8 (0.9-8.7) cells)

0.078

Marked (>10 pleomorphic cells)

9.3 (0.9-96.0)

0.061

Distribution of the large round cells often with dark nucleus at the junction

Irregular

5.0 (1.7-14.9)

0.004

Irregular distribution of the pigmentation

Present

2.3 (0.8-6.6)

0.120

Plump bright cells in big aggregation within the papillary dermis (Melanophages)

Present

0.2 (0.1-0.6)

0.005

Polycyclic papillae, as circumvolution of edged papillae (“Cord-like sign”)

Present

0.2 (0.1-0.8)

0.019

Dermal involvement

Present

25.4 (3.1-209.5)

0.003

Pleomorphic cells

Present

9.4 (2.9-30.2)

<0.001

Linear telangiectasia-like horizontal vessel

Present

8.6 (1.0-74.1)

0.049

RCM melanoma score4

≥2

11.8 (3.4-40.8)

<0.001

RCM LM score5

≥2

2.4 (0.8-7.5)

0.136

Pleomorphism (round, dendritic and other irregular shape of bright cells)

366 367 368 369 370 371 372 373 374 375 376 377

*DEJ: Dermal-epidermal junction; OR: odds ratio; CI95%: confidence interval 95%.

378 379 380 381 382 383

TABLE II: Confocal features related to melanoma in Xeroderma Pigmentosum patients. Univariate logistic regression

Characteristics

p-value

OR (CI95%)

Irregular diamond-shaped large cells (>20µ µ) in the epidermis

Scattered cells

6.6 (1.0-44.9)

0.054

Touching one another Absent

4.4 (0.3-76.5)

0.309

Presence of widespread large cells (>20µ µ) in the epidermis (pagetoid cells) Non-edged papillae Non-visible papillae Atypical cells at DEJ

Present

5.9 (1.1-30.5)

0.035

Present Present Present

3.2 (0.9-11.4)

0.078

5.3 (1.1-26.2)

0.042

Round cells with dark nucleus at the junction size

Small (= the size of epidermal cells around)

0.9 (0.1-6.2)

0.915

Large (twice the size epidermal cells around)

3.6 (0.7-19.9)

0.142

of

Follicular extension of pagetoid cells and/or atypical junctional cells around the follicular extension (adnexal spread)

Present

9.0 (1.4-56.1)

0.019

Distribution of the large round cells often with dark nucleus at the junction

Irregular

9.7 (1.9-48.3)

0.006

Distribution of the pigmentation

Irregular

5.3 (1.1-26.2)

0.042

Nest in the upper dermis RCM melanoma score4 RCM LM score5

Present ≥2 ≥2

18.4 (1.9-182.7) 5.3 (1.1-26.2) 4.4 (1.2-15.3)

0.013 0.042 0.021

384 385 386 387 388 389 390 391 392 393 394 395 396 397

*DEJ: Dermal-epidermal junction; OR: odds ratio; CI95%: confidence interval 95%.

CAPSULE SUMMARY: This was a cross-sectional study in Xeroderma pigmentosum (XP) patients with suspicious lesions submitted to confocal microscopy (RCM). We reported RCM features associated to malignancy and found that RCM has the potential to decrease the number of excisions in this high-risk population.