DOI: 10.1111/exd.12417

Original Article

www.wileyonlinelibrary.com/journal/EXD

Distinct melanoma types based on reflectance confocal microscopy Giovanni Pellacani1, Barbara De Pace1, Camilla Reggiani1, Anna Maria Cesinaro2, Giuseppe Argenziano3, Iris Zalaudek3,4, H. Peter Soyer5 and Caterina Longo3 1 Department of Dermatology, University of Modena and Reggio Emilia, Modena, Italy; 2Department of Pathology, University of Modena and Reggio Emilia, Modena, Italy; 3Skin Cancer Unit, Arcispedale Santa Maria Nuova, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS), Reggio Emilia, Italy; 4Department of Dermatology, Medical University of Graz, Graz, Austria; 5Dermatology Research Centre, The University of Queensland, School of Medicine, Translational Research Institute, Princess Alexandra Hospital, Brisbane, Qld, Australia Correspondence: Caterina Longo, MD, PhD, Dermatology and Skin Cancer Unit, Arcispedale Santa Maria Nuova-IRCCS, Viale Risorgimento, 80, 42100 Reggio Emilia, Italy, Tel.: +3905225612, Fax: +390594224271, e-mail: [email protected]

Abstract: Distinct melanoma types exist in relation to patient characteristics, tumor morphology, histopathologic aspects and genetic background. A new diagnostic imaging tool, reflectance confocal microscopy (RCM), allows in vivo analysis of a given lesion with nearly histologic resolution while offering a dynamic view of the tissue in its ‘natural’ environment. The aim of this study was to analyse cell morphology of consecutive melanomas as they appear on RCM and to correlate morphology with tumor and patient characteristics. One hundred melanomas were visualized by RCM before excision. Clinical data, confocal features and histologic criteria were analysed. Four types of melanomas were identified as follows: (i) Melanomas with a predominantly dendritic cell population (‘dendritic-cell melanomas’) typically were thin by Breslow index; (ii) Melanomas typified by roundish melanocytes were smaller in size than dendritic cell MMs, but thicker by Breslow index, and predominantly occurred in

patients with a high nevus count; (iii) Melanomas characterized by dermal nesting proliferation usually were thick by Breslow index at the time of diagnosis, although frequently smaller in size compared with the other types; and (iv) combined type melanomas may represent an evolution of dendritic cell and/or round cell types. Integration of confocal microscopy with clinical and histologic aspects may help in identifying and managing distinct tumors.

Introduction

morphology, but specific immunostaining is needed for more detailed analysis. Reflectance confocal microscopy (RCM) provides images of the skin in horizontal sections at nearly histologic resolution (15). As melanin is a high source of contrast (16), melanocytes can be easily detected (17). In addition, the horizontal sections in RCM reveal the variegated morphologies of atypical melanocytes better than haematoxylin-and-eosin- (H&E) stained sections (18–20). For this reason, RCM is a good imaging tool for in vivo exploration of melanocytic morphologies (21,22). Here, we used RCM to analyse cell morphology in a large population of melanomas and to correlate the morphologic features of the cells, either as single cells or as clusters of cells arranged in nests, with tumor and patient characteristics.

Melanoma can present with different clinical, histopathologic and biologic aspects, suggesting a heterogeneous family of diseases rather than a unique entity (1–4). Traditionally, melanomas have been classified on the basis of growth pattern and biologic behaviour as lentigo maligna (LM)/lentigo maligna melanoma (LMM), superficial spreading (SSM) and nodular melanoma (NM) (3). Other lentiginous types, such as acral lentiginous melanoma, mucosal melanoma, and rare variants have been added later to the original classification. Epidemiological data suggest that tumor types differ based on likelihood of occurrence in specific age groups or on specific anatomic sites (5–7). Genetic profiles and molecular data have been identified for each of the different types of melanoma and correlated with distinct clinical, histopathologic aspects and tumor aggressiveness (8,9). To date, the strongest prognostic factor of tumor aggressiveness remains Breslow’s thickness (10,11), although certain very thin tumors also become metastatic. Under histopathologic examination, tumor architecture (pagetoid scatter and nesting) seems to be associated with different genetic backgrounds and growth patterns (9). Moreover, different morphologic subtypes of melanocytes may also be associated with different growth patterns in melanoma (12,13). In some instances, histopathologic examination can identify morphologic features associated with aggressive growth patterns (14). Sections stained with haematoxylin and eosin can partially depict melanocytes

414

Abbreviations: RCMreflectance confocal microscopy; LMlentigo maligna; LMMlentigo maligna melanoma; SSMsuperficial spreading; NMnodular melanoma; DEJdermo-epidermal junction; SDstandard deviation. Key words: confocal microscopy – dermoscopy – histopathology – melanoma

Accepted for publication 2 April 2014

Materials and methods We retrospectively evaluated a consecutive series of 100 melanomas for which a complete set of information (clinical, dermoscopic, confocal and histologic) were available. Cases were retrieved from the database of the Department of Dermatology of the University of Modena and Reggio Emilia beginning in September 2012 and extending retrospectively until the target number was reached. The study was approved by the Institutional Review Board of Modena (protocol number 1338/CE), and the investigation was conducted in accordance to the Declaration of Helsinki. Inclusion criteria were (i) histopathologically confirmed diagnosis of melanoma; availability of: (ii) relevant clinical data in

ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Experimental Dermatology, 2014, 23, 414–418

Confocal microscopy and melanoma

patient records; (iii) dermoscopic and confocal images; and (iv) H&E-stained slides. Technical limitations of RCM prevent the laser from penetrating the skin deeply (23,24), and thus, the following lesions were excluded: (i) acral and mucosal lesions; (ii) fully ulcerated lesions; and (iii) hyperkeratotic lesions. Dermoscopic images had been acquired by means of a polarized dermoscope (DermLite Photo 3Gen LLC, San Juan Capistrano, CA, USA). Confocal imaging was performed with reflectancemode confocal laser microscope (25) (Vivascope1500; MAVIG GmBH, Munich, Germany).

Patient characteristics Demographic and clinical data [patient age, sex and Fitzpatrick skin type (26)] were retrieved. Nevus count (categorized as ‘less than 50’, ‘between 50 and 100’ and ‘more than 100’), number of clinically atypical lesions (categorized as ‘none’, ‘1 to 4’ or ‘5 or more’) and number of solar lentigines (classified as ‘absent/few’, ‘moderate’, ‘abundant’) were recorded (27–29).

Lesion aspects Lesion size was categorized as small if 10 mm. Body location was categorized as head and neck, trunk, upper or lower limb. Dermoscopic evaluation was based on 7-point check list (30). Reflectance confocal microscopy descriptors for epidermal and dermal layers were assessed, and lesions were classified according their morphology as follows (17,31): Type 1: dendritic cell melanomas. Lesions characterized by rings and/or meshwork pattern constituted by thin meshes and with a predominant population of dendritic melanocytes within the epidermis. Type 2: round cell melanomas. Lesions characterized by a meshwork pattern with large meshes, bulging nests and aggregates of predominantly large roundish cells within the epidermis Type 3: dermal nest melanomas. Lesions characterized by large aggregates of cells (cerebriform nest), invading the dermal level. Combinations of the three patterns were considered as ‘combined-type melanoma’. Lesions without any of these patterns were defined as ‘non-classifiable.’

Histopathology Tumor stage according to AJCC was reported, and tumor thickness measured according to Breslow index was divided into the following categories: (i) in situ tumors; (ii) thin invasive melanoma, >0.1 and 0.5 and 1 mm. The presence of 1 or more mitoses per mm (2), ulceration and pattern of growth were recorded. Tumors were classified as LM/LMM, SSM or NM.

Statistics Absolute and relative frequencies were calculated for all parameters associated with the different tumor categories. Mean, standard deviation (SD), median and interquartile ranges were calculated for continuous parameters. Pearson’s chi-square test was used to compare differences between different categories. Mann–Whitney U-test was calculated for continuous measures. A P value

Distinct melanoma types based on reflectance confocal microscopy.

Distinct melanoma types exist in relation to patient characteristics, tumor morphology, histopathologic aspects and genetic background. A new diagnost...
652KB Sizes 4 Downloads 4 Views