Cutaneous Oncology: from Research to Diagnosis and Management

BioMed Research International

Cutaneous Oncology: From Research to Diagnosis and Management

Guest Editors: Ieva Saulite, Elisabeth Roider, Razvigor Darlenski, Ahmad Jalili, and Emmanuella Guenova Cutaneous Oncology: From Research to Diagnosis and Management BioMed Research International Cutaneous Oncology: From Research to Diagnosis and Management

This is a special issue published in “BioMed Research International.” All articles are open access articles distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Contents

Cutaneous Oncology: From Research to Diagnosis and Management Ieva Saulite, Elisabeth Roider, Razvigor Darlenksi, Ahmad Jalili, and Emmanuella Guenova Volume 2016, Article ID 7819428, 2 pages

Sézary Syndrome and Atopic Dermatitis: Comparison of Immunological Aspects and Targets Ieva Saulite, Wolfram Hoetzenecker, Stephan Weidinger, Antonio Cozzio, Emmanuella Guenova, and Ulrike Wehkamp Volume 2016, Article ID 9717530, 15 pages

The Skin as an Early Expression of Malignancies in the Neonatal Age: A Review of the Literature and a Case Series Vito Mondì, Fiammetta Piersigilli, Guglielmo Salvatori, and Cinzia Auriti Volume2015,ArticleID809406,10pages

Surgical Treatment and Prognosis of Angiosarcoma of the Scalp: A Retrospective Analysis of 14 Patients in a Single Institution Jun Ho Choi, Kyung Chan Ahn, Hak Chang, Kyung Won Minn, Ung Sik Jin, and Byung Jun Kim Volume 2015, Article ID 321896, 8 pages

Computer-Aided Decision Support for Melanoma Detection Applied on Melanocytic and Nonmelanocytic Skin Lesions: A Comparison of Two Systems Based on Automatic Analysis of Dermoscopic Images Kajsa Møllersen, Herbert Kirchesch, Maciel Zortea, Thomas R. Schopf, Kristian Hindberg, and Fred Godtliebsen Volume 2015, Article ID 579282, 8 pages

Wide Local Excision for Dermatofibrosarcoma Protuberans: A Single-Center Series of 90 Patients Byung Jun Kim, Hyeonwoo Kim, Ung Sik Jin, Kyung Won Minn, and Hak Chang Volume2015,ArticleID642549,9pages

Comparison of the Clinical Characteristics and Outcome of Benign and Malignant Eyelid Tumors: An Analysis of 4521 Eyelid Tumors in a Tertiary Medical Center Yu-Yun Huang, Wen-Yih Liang, Chieh-Chih Tsai, Shu-Ching Kao, Wei-Kuang Yu, Hui-Chuan Kau, and Catherine Jui-Ling Liu Volume 2015, Article ID 453091, 5 pages

+ Single-Fraction Radiotherapy for CD30 Lymphoproliferative Disorders Michelle S. Gentile, Maria Estela Martinez-Escala, Tarita O. Thomas, Joan Guitart, Steven Rosen, Timothy Kuzel, and Bharat B. Mittal Volume2015,ArticleID629587,10pages Hindawi Publishing Corporation BioMed Research International Volume 2016, Article ID 7819428, 2 pages http://dx.doi.org/10.1155/2016/7819428

Editorial Cutaneous Oncology: From Research to Diagnosis and Management

Ieva Saulite,1 Elisabeth Roider,2 Razvigor Darlenksi,3,4 Ahmad Jalili,5 and Emmanuella Guenova6

1 Department of Infectology and Dermatology, Riga Stradins University, Riga, Latvia 2Department of Dermatology, Cutaneous Biology Research Center, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA 3Department of Dermatology and Venereology, Tokuda Hospital Sofia, Sofia, Bulgaria 4Section of Dermatology and Venereology, Trakia University, Stara Zagora, Bulgaria 5Division of Immunology, Allergy and Infectious Diseases (DIAID), Department of Dermatology, Medical University of Vienna, Vienna, Austria 6Department of Dermatology, University Hospital of Zurich, University of Zurich, Gloriastrasse 31, 8091 Zurich, Switzerland

Correspondence should be addressed to Emmanuella Guenova; [email protected]

Received 18 April 2016; Accepted 19 April 2016

Copyright © 2016 Ieva Saulite et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Skin cancer incidence rates show a tendency for continuous eyelid tumors. Furthermore applied treatment method and increase. Along with this rise, cutaneous oncology maintains outcome in case of benign and malignant eyelid tumors are an extensive interest for clinicians and researchers as well evaluated aiming to support clinical decision to reduce the as the pharmaceutical industry. This special issue highlights esthetic disfigurement and morbidity especially in case of and discusses exciting implications in exploration of broad malignant eyelid tumors. spectrum of rare and common conditions with respect to Improving aspects of early diagnosis of skin cancer both diagnosis and treatment of cutaneous malignancies. remains a fundamental concern. The role of computer assis- An important aspect concerning angiosarcoma, a highly tance for clinical decision with regard to detecting skin aggressive malignant tumor that still lacks a common consen- cancer needs to be defined and evidenced by appropriate sus for the treatment approach, has been targeted in the article studies. A profoundly actual up-to-date study “Computer- titled “Surgical Treatment and Prognosis of Angiosarcoma of Aided Decision Support for Melanoma Detection Applied theScalp:ARetrospectiveAnalysisof14PatientsinaSingle on Melanocytic and Nonmelanocytic Skin Lesions: A Com- Institution” by J. H. Choi et al. Although authors report a poor parison of Two Systems Based on Automatic Analysis of prognosis of angiosarcoma despite aggressive treatments, Dermoscopic Images” by K. Møllersen et al. compares capa- they demonstrate that in case of angiosarcoma sufficiently bility of two clinical decision support computer systems to deep resection margins are essential for controlling local detect not only melanocytic but also nonmelanocytic skin recurrence, further highlighting the need for studies aiming lesions. Authors report potential of commercially available to evaluate multidisciplinary treatment modalities. system Nevus Doctor being able to detect nonmelanocytic A well-structured study of a high clinical relevance skin cancer along with remaining melanoma sensitivity “Comparison of the Clinical Characteristics and Outcome of and current limitations due to misclassification of benign Benign and Malignant Eyelid Tumors: An Analysis of 4521 nonmelanocytic lesions. Eyelid Tumors in a Tertiary Medical Center” by Y.-Y. Huang CD30+ lymphoproliferative disorder (LPD) is a rare et al. provides essential demographic and clinical features variantofT-celllymphomathatinsomecasesisrecurrentand to support differentiation between benign and malignant treatment resistant. Therefore for palliation of CD30+LPD 2 BioMed Research International to avoid unnecessary side effects of aggressive treatment approaches radiotherapy may be considered. However there is a lack of data endorsing radiotherapy (RT) in CD30+ LPD. M. S. Gentile et al. in the study titled “Single-Fraction Radiotherapy for CD30+ Lymphoproliferative Disorders” in a retrospective review prominently assess the clinical response of refractory or recurrent CD30+ lymphoproliferative disorder lesions to palliative radiation therapy suggesting palliative localized RT for symptomatic CD30+ LPD refractory or recurrent to other therapies. Study supports that palliative radiation therapy is an effective well tolerated treatment modality for CD30+ LPD refractory or recurrent to other therapies and reports single fraction of 750–800 cGy being as effective as a multifractionated course and more convenient. In another rare malignancy, low-grade sarcoma, dermatofibrosarcoma protuberans (DFSP) margins of the tumor tend to extend beyond the macroscopic ones. Therefore the incomplete removal of the tumor leads to recurrences. This concern is accurately aimed in comprehensive clinical study titled “Wide Local Excision for Dermatofibrosarcoma Protu- berans: A Single-Center Series of 90 Patients” by B. J. Kim et al. which evaluates efficacy of wide local excision for DFSP. Authors report that wide local excision with adequate lateral and deep margins can effectively control local recurrence rate and is a simple and effective method to treat DFSP. Although most skin rashes in the neonatal age are benign, some may represent malignancy or be a feature related to other abnormalities. Therefore to provide the best possible outcome a considerable assessment and an adequate follow- up for each condition are of high value. A profoundly written educational review and a case series “The Skin as an Early Expression of Malignancies in the Neonatal Age: A Review of the Literature and a Case Series” by V. Mond`ıetal.com- prehensively discuss the main tumor types presenting with a cutaneous involvement in neonates, additionally providing a collection of didactic cases presenting with an early skin manifestation of malignancies in illustrative case series. A review article titled “Sezary-Syndrome´ and Atopic Der- matitis: Comparison of Immunological Aspects and Targets” by I. Saulite et al. provides an interesting perspective by comparing currently accepted immunological aspects and possible therapeutic targets in atopic dermatitis (AD) and Sezary´ syndrome (SS). Authors discuss clinical appearance and the hallmark immunological characteristics of SS highlighting some striking similarities with acute flares of atopic dermatitis (AD). Taking into consideration the overlap of several immunological features, the application of similar therapeutic approaches in certain stages of both diseases that may come into consideration is discussed. Hence, with a definite commitment to improve every aspect contributing to skin cancer this special issue consisting of original articles and educational reviews aims to be trans- lated in clinical advancements for patients with skin cancer. Ieva Saulite Elisabeth Roider Razvigor Darlenksi Ahmad Jalili Emmanuella Guenova Hindawi Publishing Corporation BioMed Research International Volume 2016, Article ID 9717530, 15 pages http://dx.doi.org/10.1155/2016/9717530

Review Article Sézary Syndrome and Atopic Dermatitis: Comparison of Immunological Aspects and Targets

Ieva Saulite,1,2 Wolfram Hoetzenecker,1 Stephan Weidinger,3 Antonio Cozzio,1 Emmanuella Guenova,1 and Ulrike Wehkamp3

1 Department of Dermatology, University Hospital of Zurich, University of Zurich, Gloriastrasse 31, 8091 Zurich, Switzerland 2Department of Infectology and Dermatology, Riga Stradins University, Riga, Latvia 3Department of Dermatology, Christian-Albrechts-University Kiel, Kiel, Germany

Correspondence should be addressed to Emmanuella Guenova; [email protected]

Received 4 December 2015; Accepted 30 March 2016

Academic Editor: Arjen F. Nikkels

Sezary´ syndrome (SS), an aggressive form of erythrodermic pruritic cutaneous T cell lymphoma (CTCL), from an immunological perspective characterized by increased Th2 cytokine levels, elevated serum IgE and impaired cellular immunity. Not only the clinical appearance but also the hallmark immunological characteristics of SS often share striking similarities with acute flares of atopic dermatitis (AD), a common benign chronic inflammatory skin disease. Given the overlap of several immunological features, the application of similar or even identical therapeutic approaches in certain stages of both diseases may come into consideration. The aim of this review is to compare currently accepted immunological aspects and possible therapeutic targets in AD and SS.

1. Introduction inflammatory skin disease with a lifetime prevalence of 15– 20% in developed countries [10]. The majority of patients Sezary´ syndrome (SS) is a rare erythrodermic and leukemic show an onset in early childhood and a remission until ado- variant of cutaneous T cell lymphomas (CTCL) that belongs lescence. However, recent prevalence estimates in adults of up to the heterogeneous group of extranodal non-Hodgkin’s to 10% indicate that the rate of persistent and/or adult-onset lymphomas (NHL) arising from the malignant proliferation disease is higher than previously assumed [11, 12]. AD is an of skin-homing T cells [1, 2]. SS together with mycosis important differential diagnosis of SS in adults with erythro- fungoides (MF) are the most common forms of CTCL dermic dermatitis [10]. Although in majority of cases there accounting for around 65% of cases whereas SS represent are characteristics such as typical predilection sites for AD around 3% of all CTCL [3]. CTCL are assumed to have a male and palmoplantar hyperkeratosis for SS that allow clinically predominance and the median age at onset of the disease is distinguishing between AD and E-CTCL, in some excep- between the fifth and sixth decade [4, 5]. tional cases of erythroderma especially among the elderly The behaviour of the SS is aggressive with a median population initially it might be a clinical challenge to define survival of 1–5 years [3, 6, 7]. SS and erythrodermic MF thediagnosis.Thecomparableclinicalfeaturesarefurther (E-MF), which is considered to be an advanced form of reflected by some overlapping immunological peculiarities, MF with absent or minimal blood involvement, may be in particular an epidermal barrier deficiency, and a cutaneous referred to as erythrodermic CTCL (E-CTCL) [3, 8]. If blood infiltration by CD4+ T helper cells expressing the skin- involvement is present, the term leukemic CTCL (L-CTCL) homing receptor cutaneous lymphocyte-associated antigen is used and therefore it is applicable to every case of SS (CLA) and chemokine receptor 4 (CCR4). Interestingly, both [1, 2]. Besides due to the lack of clear diagnostic markers the AD and SS show increased production of Th2 cytokines such differential diagnosis of various erythrodermic skin diseases as interleukin 4 (IL-4), interleukin 5 (IL-5), and interleukin is still challenging [9]. Atopic dermatitis is a common chronic 13 (IL-13) as well as CCR4-binding chemokines that is 2 BioMed Research International

(a) (b)

Figure 1: Clinical appearance of patient with Sezary´ syndrome (a) and atopic dermatitis (b). characteristic also of the acute phase of AD [13–15]. As a varying from discoloration to subungual hyperkeratosis and consequence of the epidermal barrier deficiency and the ocular involvement, most frequently eyelid ectropion [20– diminished Th1 and Th17 cell immunity, the skin ofAD 22]. Elderly patients with erythrodermic eczematous pruritic patients shows a less diverse surface microbiome and an skin may be a great clinical challenge for physicians with increased susceptibility towards cutaneous colonization and regard to differential diagnosis. Some case reports have infection with Staphylococcus aureus (S. aureus)[10,15].The described SS arising in patients with a long history of AD [23– skin microbiome has not yet been systematically examined 26]. However, a study showed no significant difference in the in CTCL, but there are preliminary data indicating increased prevalence of atopy in SS compared to MF and the general S. aureus colonization rates in MF and SS [16]. Both AD population [23]. and SS benefit from topical barrier restoring and rather To confirm the definite diagnosis in E-CTCL, clin- unspecific topical or systemic immunosuppressive treatment, icopathological correlation often including multiple skin although SS often shows slower and/or weaker responses biopsies with histopathological and immunohistochemical [10, 17]. As insights into the precise molecular mechanisms investigations and in most cases staging examinations (blood, and key immunological networks driving inflammation grow, lymph node, and other organs) are necessary [9, 27–29]. summarizing the knowledge about immune responses in these Th2 cell-dominated diseases may potentially allow 3. Clinical Characteristics of AD drawing conclusions about different markers and therapeutic targets in both of the diseases. The aim of this review is to The most characteristic features of AD are intense itch and compare the immunological aspects and therapeutic targets recurrent eczematous skin lesions, which typically show an in AD and CTCL. age-related morphology and distribution. Infants most often present with eczematous skin lesions on the cheeks and the 2. Clinical Characteristics of E-CTCL scalp while in childhood predilection sites are the flexures, neck, and dorsal aspects of the limbs (Figure 1(b)). Starting SS is defined by a typical clinical triad consisting of erythro- with adolescence flexural areas, head and neck, shoulders, derma, peripheral lymphadenopathy, and peripheral blood and hands are predominantly affected [30, 31]. The majority involvement. Although in the majority of SS cases rapid of patients display generalized skin dryness and a personal onset of the clinical manifestations can be observed, in some or familial history of atopy. Other associated features are patients a long medical history including disabling pruritus a hyperlinearity of the palms and soles, Dennie-Morgan as well as nonspecific dermatitis is present. Cutaneous man- infraorbital folds, and Hertoghe’s sign [10]. In general, AD ifestations in E-CTCL comprise a broad clinical spectrum shows a wide spectrum of clinical features and trajectories, varying from mild erythema to generalized exfoliative ery- in particular in adulthood, where the disease often presents throderma complicated by electrolyte dysregulation and high as eczematous erythroderma and appears to show a male output cardiac failure due to the extensively dilated skin ves- predominance [31, 32]. Adult-onset AD often shows an sels [18, 19] (Figure 1(a)). Erythroderma is often accompanied untypical distribution and morphology of lesions and is not by severe pruritus. Additionally, the patients may present with well captured by classical diagnostic criteria and appears to be palmoplantar keratoderma and alopecia and nail changes less closely associated with atopy as compared to early-onset BioMed Research International 3

CD8+ T cell DC IFN-𝛾 DC IFN-𝛼 IL-12 Th22 cell

IL-4 IL-4 IL-5 B cell Naive IL-13 Th2 cell IL-10 Th cell Skin-draining Malignant CD4+ Eo lymph node T cell Skin-draining lymph node Th2>Th1 Th2>Th1

IgE Blood vessel Epidermis

(a) Sezarý syndrome/erythrodermic mycosis fungoides (b) Acute atopic dermatitis

Figure 2: Th2 prevailing immune responseezary inS´ syndrome (SS)/erythrodermic MF (E-MF) and atopic dermatitis (AD). (a) SS/E-MF with malignant T cells circulating in the blood promote an immune response with production of Th2 phenotypic cytokines. Therefore suppressing the Th1 immune properties leads to impaired host immune response.) (b In the acute phase of atopic dermatitis naive T helper cells are primed in Th2 cells under the influence of activated skin resident DC which have the capacity of migrating to skin-draining lymph nodes. DC:dendriticcell,Eo:eosinophilicgranulocyte. forms, although reliable clinical and epidemiological data are antitumor response promoted by dendritic cells [35–38]; on lacking [10]. the contrary, the disease progression of MF was found to be associated with a remarkable decrease of normal Treg and cytotoxic T cells [39]. The exact role of Tregs in CTCL 4. Immunological Aspects still remains controversial. Malignant T cells which, in some SS patients, exhibit phenotype characteristics for Tregs may In both L-CTCL and acute AD there is a predominance of contribute to the downregulation of the local antitumor Th2 immune response [33, 34]. An overview is shown in response as well as systemic immunosuppression [37, 39, Figure 2. The complexity of the T cell compartment is not fully 40]. Furthermore, it has been suggested that SS might be a understood,andtheroleofTcellsintheregulationofchronic malignancy of FOXP3+ (forkhead box P3) regulatory T cells, inflammatory processes in skin diseases as well as in cancer Th2cells,andTh17,whereasthepresenceofTh17inanother is puzzling and yet to be defined. study could not be confirmed [33, 41–43]. The pathophysiological mechanisms underlying both Several studies of peripheral blood mononuclear cells entities, the one being a benign inflammatory skin disease, have demonstrated increased production of Th2 cytokines with the other representing a hematological malignancy of and reduced production of Th1 cytokines in L-CTCL [44, the skin, will be described with a focus to enlighten the 45]. Gene expression studies confirmed elevation of Th2 similarities and differences for both conditions. associatedgenesinthePBMCandskinofpatientswithL- CTCL [46, 47]. The association of increased IL-4 production 4.1. Distinct Features of Pathogenesis of SS (L-CTCL) and AD with advanced stage L-CTCL suggests that the source of excess Th2 cytokines may be the malignant clone itself. 4.1.1. Malignant T Cells in CTCL. Acrosstalkbetween This hypothesis has recently been corroborated by study- different cells of the innate immune system, benign bystander ing the functional bias of malignant T cells in L-CTCL T cells, and malignant T cells is crucial for the immune patients in whom the malignant T cell clone could be responses in CTCL. conclusively identified by staining with antibodies against In initial MF skin lesions, an increased number of CD8 TCR V 𝛽. Flow cytometrical assessment of the intracellular cytotoxic cells, in addition to FOXP3 regulatory T cells cytokine production demonstrated that both the malignant (Tregs), have been detected, which was attributed to an clone and surprisingly the remaining benign T cells expressed 4 BioMed Research International high levels of the Th2 cytokines IL-4 and IL-13 and negligible might be induced by TLR2 ligands through IL-4-mediated levels of IFN-𝛾 upon activation [33]. suppression of IL-10 [63]. Several recent studies were dedicated to the origin of the malignant T cells. Interestingly, malignant T cells, isolated 4.1.3. Factors Contributing to Impaired Cellular Immunity fromtheskinofSSpatients,showednotonlyCCR7and in CTCL. In addition, many other mechanisms might be CCR4 but also high expression of L-selectin and CD27, a responsible for the inhibition of cellular immunity by phenotype characteristic of central memory T cells [1]. On the CTCL such as dysregulated expression of the immunoreg- other hand absence of CCR7/L-selectin and CD27 expression ulatory proteins (e.g., CTLA-4, PD-L1 (programmed-death- has been shown in T cells from MF lesions, whereas CCR4 ligand 1), and Fas ligand (Fas L)) and constitutive acti- and CLA are highly expressed suggesting a phenotype of skin vation of Jak/Stat pathway, which promotes transforming resident effector memory T cells [1]. These findings ledto growth factor 𝛽 (TGF-𝛽) and IL-10 secretion [43, 64–66]. thehypothesisthatSSisamalignancyarisingfromcentral Programmed-death-receptor-1 (PD-1), a membrane molecule memory T cells, while MF is a malignancy of skin resident of CD28/CTLA-1 receptor family, plays an important role in effector memory T cells, supporting evidence that SS and cellular immunity. By interaction with its ligands it has been MF should be considered as separate lymphomas originating shown to inhibit T cell activation and proliferation [67]. It from distinct functional T cell subsets [1]. has been demonstrated to be highly expressed by neoplastic In 2004 Rubben¨ et al. proposed the intriguing hypothesis T cells in SS [68]. Although its role in pathogenesis of CTCL that CTCL arises from genetically unstable subclones that is not clear, it might contribute to the immunosuppression in undergo a multilineage progress into a stable clone leading to SS [68, 69]. subsequent proliferation of neoplastic T cell population that Defective apoptosis has been shown to be characteristic of originates MF [48]. SS. It has been demonstrated to correlate with decreased or In general, progression of the disease correlates with a impaired death receptor mainly Fas expression by neoplastic decrease of the T cell receptor (TCR) repertoire, paralleled by T cells, resulting in perpetual neoplastic T cell proliferation a predominance of clonal, malignant CD4+ T cell population, [70–75]. expressing a single TCR clone [49]. Further it has been shown that adhesion molecules and 4.1.2. T Cells in AD. Regarding immune dysregulation in AD, chemokines have a significant contribution to skin-homing itisknownthatinacutephaseofADadenseinfiltrateof of malignant T cells in CTCL. The mechanisms are not CD4+ cells as well as allergen specific CD4+ and CD8+ T yet fully elucidated, but some factors that promote T cells cells can be found in the affected skin lesions, thus promoting prone to skin-homing, for example, adhesion molecules cutaneous inflammation [50]. and chemokines, have been identified. Among them are During an acute onset of the disease Th2 bias is charac- cutaneous lymphocyte-associated antigen (CLA), chemokine teristic whereas along with the chronification an increasing receptor 4 (CCR4), CCR10, and CCR7 whose expression has proportion of the skin is being infiltrated by Th1 cells been demonstrated by malignant T cells in patients with MF accompanied by Th2, Th0, and Th22 cells [13, 51–54]. andSS[76–82].CXCR4,areceptorforCXCL12,whichcan In AD naive T helper cells are being polarized into Th2 also be observed in SS, may contribute to cell skin recruitment phenotype by activated skin resident dendritic cells that have andaccumulationthroughtheregulatoryactivityofCD26in migrated to local lymph nodes [52, 55]. CTCL [83, 84]. The mechanisms are not yet fully elucidated, The hallmark Th2 cytokines IL-4 and IL-13 play important but some factors that promote T cells prone to skin-homing, role in the pathogenesis of AD. Following IL-4 dependent for example, adhesion molecules and chemokines, have been induction of IgG class switch in B-cells, a subsequent eleva- identified [77, 82, 83, 85]. tion of IgE levels can be frequently observed in AD patients [56]. Furthermore, IL-4 is important for the functional 4.1.4. Aspects Leading to Impaired Skin Barrier in AD. The phenotypeofimmigratingDCprecursorsandtherebyheavily major hallmarks in the pathogenesis of AD are an impaired influences the phenotype of an ongoing immune response epidermal skin barrier function and an immune dysregula- [34]. tion. Genetic background with filaggrin protein gene muta- Thymus and activation regulated chemokine (TARC/ tions along with other factors such as skin cytokine imbalance CCL17), a member of the Th2 chemokine family, serum level leads to decreased filaggrin expression, which belongs to one has been suggested to be a useful clinical biomarker for AD of the most crucial factors underlying the epidermal barrier treatment and disease severity [57, 58], whereas in CTCL dysfunction in AD [10, 31]. Moreover, filaggrin deficiency at disease progression chemokine receptors, expressed by has been associated with subclinical inflammation, reduced skin-infiltrating T cells and surface molecules, have shown resistance to irritants and haptens, and an enhanced percu- the tendency to decrease due to loss of these markers taneous allergen sensitizing [10]. Another peculiarity in AD subsequently followed by diminished epidermotropism [59, is decrease and alterations in lipids such as ceramides of the 60]. stratum corneum. Besides keratinocytes not only maintain However in contrast for chronic AD lesions there is an the first line of physical barrier but also express pattern increase of Th1 cytokines: interferon-𝛾, IL-12, IL-5, and GM- recognition receptors for several agents including proteolytic CSF [61, 62]. Recently, it has been suggested that in Th2 allergens that are also capable of inducing Th2 cell mediated cell mediated dermatitis persistence of chronic inflammation immune responses [10]. BioMed Research International 5

Several host and environmental aspects contribute to 4.3. Joint Aspects of Pruritus in AD and SS (L-CTCL). the epidermal barrier dysfunction in AD that increase skin Several receptors, secreted molecules (histamine, nerve penetration for allergens, microorganisms, and irritants. growth factor (NGF), and substance P (SP)), proteases, and cytokines/chemokines (thymic stromal lymphopoietin (TSLP), IL-2, IL-4, IL-13, and IL-31) along with many oth- 4.2. Joint Aspects of Pathogenesis of SS (L-CTCL) and AD ers are described as contributing to chronic pruritus [96]. Although the role of distinct players in AD is ambiguous, 4.2.1. Th2 Weighted Immune Response. Although in early IL-31 strongly contributes to pruritus in AD also correlating stages of MF a dominance of Th1 immune answer is pre- with the severity of the disease [97, 98]. Interestingly, some vailing, a later progression of the disease reveals an immune studies have suggested that there is no general relation- response that is dominated by Th2 malignant T cells. Th2 ship between IL-31 protein expression and pruritus in Th2 weighed immune response is characteristic of an acute phase weighted diseases. Therefore, IL-31 might play a distinct role of atopic dermatitis as well [18, 30, 33, 44]. in the pathogenesis of AD [17]. In SS, a Th2 weighed immune response is present with Although tumor cells in MF/SS exhibit similar cytokine an overproduction of the typical cytokine profile, IL-4, IL-5, profile with AD and both are characterized with pruritus, IL-10, and IL-13, respectively, and additionally elevated levels the analysis of the IL-31 pathway in MF/SS patients regarding of serum immunoglobulin E (IgE) and immunoglobulin A serum levels as well as receptor expression does not suggest (IgA) and peripheral eosinophilia [33, 42, 44, 86]. Along a central role of IL-31 in MF/SS pathogenesis. Nevertheless with the disease progression, a decline in the number and increased levels of IL-31 were observed in patients with severe activity of benign immune cells results in an impaired cell pruritus; therefore it might be a rationale for therapeutic mediated cytotoxicity and decreased antigen-specific T cell approach for some patients [99]. responses and consecutively in a severe immunodeficiency [17, 33]. Interestingly, in SS patients a global Th2 bias with 4.4. Joint Factors of Immune Dysregulation Leading to enhanced production of Th2 cytokines has been shown to Increased Infectious Susceptibility in SS and AD. Regarding be characteristic of both benign and malignant T cells. Th2 atopic dermatitis it is still a matter of debate whether the bias was demonstrated to be intrinsic in malignant T cells but primary factor is the disturbed immune response that results extrinsic in benign T cells. It has been demonstrated that the with a defective epidermal skin barrier function or vice versa. Th2cytokinesfrommalignantcellsarecapableofinhibiting Nevertheless, there is no doubt that both components have the Th1 responses [33]. a strong contribution and there is a close interaction among Atopic Dermatitis Is Also Th2 Prototypic Disease in the them; moreover once the inflammatory cascade has been Acute Phase. Consequently, in the acute phase a predomi- activatedtheybothbelongtotheCirculus vitiosus. nance of Th2 cytokines, IL-4, IL-5, and IL-13, is observed; Keeping in mind that acute AD and L-CTCL are both therefore with regard to cytokine profile some parallels characterized by Th2 phenotypic immune response one between AD and L-CTCL can be drawn. would expect some overlap in clinical consequences as well. This Th2 phenotype shift may be an important factor In both AD and L-CTCL Th2 cytokines modulate the prompting an infectious susceptibility observed in patients immune response and lead to defective cutaneous barrier with SS as well as AD [14, 75, 76]. A better understanding function by impairing keratinocyte protein differentiation of these interactions in SS and AD would promote further and downregulating the antimicrobial peptides in the skin targets for treatment. and therefore the innate immunity of the skin. Both the immune dysregulation and the decreased skin barrier predis- 4.2.2. Overlapping Factors That May Promote Th2 Weighed pose to an extensive bacterial colonization and increased risk Immune Response in Both AD and SS. Eotaxins, chemokine of skin infections, which can be observed in patients with AD ligands (CCL) CCL11 and CCL26 that are expressed by and SS as well [100–102]. epidermal keratinocytes and dermal fibroblasts support the Whereas for both AD and L-CTCL the immune dysregu- chemotaxis of Th2 cells positive for chemokine receptor 3 lation promotes the increased susceptibility to infections, in (CCR3) and eosinophils, thereby also promoting the Th2 bias case of L-CTCL this aspect appears to play the major role. [39, 87–90]. There is an increased expression of CCL11 and As already mentioned above, there is reduced skin CCL26bothinserumandinlesionalskininADpatients microbiome diversity among AD patients with a prevailing therefore suggesting that it might also play an important role colonization of Staphylococcus aureus (S. aureus). In patients in the pathogenesis of the disease [91, 92]. Also in patients with MF and SS, there likewise seems to be an overabundance with advanced MF/SS elevated CCL11 and CCL26 levels in of staphylococcal carriage to a similar extent as in AD [16, comparison with healthy individuals have been demonstrated 103–105]. [90]. Besides IgE antibodies against S. aureus toxins have been In addition, elevated serum levels of TSLP that activates shown to exhibit superantigen properties, which appear to immature myeloid dendritic cells (DC) to produce CCL17 in correlate with the severity of the disease [106]. In addition, CTCL patients were detected. Also in AD high expression S. aureus toxins themselves exhibit superantigen properties in keratinocytes in affected skin lesions was seen and TSLP [107]. Moreover, it has been shown that staphylococcal might promote leading towards a Th2 phenotypic immune enterotoxin B (SEB) upregulates IL-31 in peripheral blood response [55, 93–95]. mononuclear cells [108, 109]. It is assumed that signal 6 BioMed Research International transducer and activator of transcription 3 (Stat3) and the system. The cell of origin is a central memory T cell with immunoregulatory cytokine interleukin 10 (IL-10) may play the ability of circulation in skin, blood, and lymph nodes [1]. an important role in immune dysregulation in CTCL. During Therefore, targeted therapy for SS always means interacting disease progression malignant activation of Stat3 and expres- with immune mechanisms. sion of IL-10 increase in parallel with the evolving immune In AD multiple factors seem to contribute to this con- impairment [42, 65, 110, 111]. dition such as underlying genetic predisposition and envi- Furthermore, it has been shown recently that microbial ronmental factors. Excessive T cell activation is characteristic toxins, namely, staphylococcal enterotoxins, might be part of of AD with still unclear exact pathophysiologic mechanisms vicious circle not only as an epiphenomenon but also by stim- [122]. Targeting the T cells and/or the environment of ulating benign T cells to induce activation of the immunoreg- neoplasm/inflammation is a reasonable approach in both dis- ulatory Stat3/IL-10 axis in malignant T cells. Accordingly in eases. The aim-result of targeted therapy is either induction CTCL colonization with S. aureus that produces enterotoxin of immunotoxic effects on cancer cells or modification of may promote the skin immune dysregulation [112]. immunological mechanisms. The latter is the more common Eradication of pathogen skin microflora has been asso- in the currently developed agents. In malignant T cell prolif- ciated with clinical improvement by several authors [16, 104, eration blocking immunosuppression or enhancing the local 112, 113]. immune response is supposed to be beneficial. Many targets Furthermore, therapies that inhibit Th2 cytokine activ- to act on these differing approaches are under exploration. ity, hence keeping a balance of Th1 responses, may have the potential to enhance both antipathogen and antitumor 5.2. Targets on T Cells. Alemtuzumab is a humanized mon- responses [16, 33, 114, 115]. oclonal anti-CD52 directed antibody. CD52 is described as a pan-T- and B-cell-marker, expressed also on dendritic cells 5. Therapy andmacrophages.Alemtuzumabhasproventobeespecially useful in SS and MF with leukemic involvement, which seems In both SS and AD there is a repertoire of standard therapies to be related to the cell of origin of the clonal malignancy and that is used according to current guidelines in respect of its surface marker profile [1, 123]. It leads to a depletion of the extent of skin and systemic involvement [116–119]. These circulating lymphocytes and thereby suppression of immune standards will not be discussed in this review. In the past responses [124]. Application may result in severe adverse years newly developed therapeutic agents have led to a events, mainly infections [125]. Its use in AD has to be dramatically improved therapeutic outcome and change of discussed carefully. There is no data published on this topic. treatment rationale for many diseases. This is particularly Ipilimumab (Yervoy5) is a monoclonal antibody targeted true for cancer therapy and in dermatology revolutionizing against CTLA-4. CTLA-4 obviously is overexpressed in the treatment for melanoma implementing targeted therapy. CTCL and SS and coincides in the latter with abnormal The aim of targeted therapy is the destruction of tumor findings for IL-10 and Foxp3 [126]. This phenotype would cells and induction of as few side effects as possible. Due suggest a differentiation towards Tregs [41]. Blocking CTLA- to the complexity of cancer and having no exclusive target 4 might therefore be an interesting therapeutic approach. markers on cancer cells this concept comes not always However, there is no data yet on CTLA-4 inhibition in true but nonetheless has led to an immense advancement CTCL and the exact mechanism of action has not yet been and better understanding of the pathophysiology in many completely elucidated. Overexpression of CTLA-4 in CTCL diseases. Targeted therapies have been developed not only couldbeinterpretedasfeatureeitherofanimmuneescape for cancer but also for inflammatory diseases. Despite the mechanism or potentially of immunosuppression [127]. In efforts in developing effective treatments in CTCL until today, Hodgkin lymphoma a study for exploration of the combina- apart from stem cell transplantation cure is not achievable tional therapy of ipilimumab with brentuximab vedotin and [120]. However, induction of partial or complete remission is nivolumab has been initiated. The three agents are used in possible [121]. three different arms comparing nivolumab + brentuximab In the following, we focus on immunologic checkpoints vedotin versus ipilimumab + brentuximab vedotin versus the that are of interest in SS and are currently implemented in combination of all the three drugs. Transferring the results clinical studies. We compare the checkpoints to data drawn of this study will be highly interesting for the treatment of from studies in AD and/or evaluate potential benefits from CTCL, especially CTCL with CD30 expression. transferring therapeutic approaches from one entity to the PD-1 and programmed-death-ligand 1 (PD-L1) are other. The list of targets mentioned here is not exhaustive. immunologic checkpoints on T cells and cancer cells, All information about ongoing studies without published respectively, and their therapeutic blockage has led to results was obtained at the official homepage of the registry convincing results for overall survival improvement in non- and results database for clinical studies in human par- small-cell lung cancer, melanoma, and other solid tumors ticipants: https://clinicaltrials.gov/. An overview about the [128]. Approved in melanoma and non-small-cell lung cancer targets discussed in this review is given in Table 1. Detailed in Europe are two PD-1-inhibitors pembrolizumab (MK- information and references are to be found in the text. 3475/Keytruda5) and nivolumab (Opdivo5).Aphase2study with pembrolizumab for the treatment of relapsed/refractory 5.1. Targeting Structures of the Immune System. SS arises from MF/SS is active. Estimated primary completion date is the lymphocytic system, which is per se part of the immune January 2018. PD-L1-inhibitors are not yet approved but BioMed Research International 7 ’s lymphoma, AML: acute myeloblastic leukemia, In phase 2 in NHL, interestingopinions in for CTCL, use controversial in AD Not likely to be used innivolumab, AD, ipilimumab combination study & with brentuximab inIn HL phase ongoing 1 in HL, interesting in SS/CTCL Topical application, leading to inflammatory reactions, not promising for AD Interesting because of subcutaneousrestriction to application skin (no surface) Has been studied in NHL inavailable phase 1, data not yet Interesting as well in CTCL Had been withdrawn from the marketnewly for approved CTCL for and multiple sclerosis (noreasons), medical not likely to beInteresting used in CTCL, in less AD in AD,with combination nivolumab, study ipilimumab, and brentuximab in HL ongoing Used in AML, interesting for SS/CTCL Both in malignant melanoma Basal cell carcinoma, actinic keratoses, and genital warts Malignant melanoma Approved for (in Europe) Comments (more information & references in text) X X X X Indication AD ´ ezary syndrome/cutaneous T cell lymphoma and atopic dermatitis in alphabetical order. X XMultiplesclerosis XCTCLNotlikelytobeusedinAD SS/CTCL Indication Full names of targets in the text. ∗ NCT02243579 X NCT00043420 X NCT02347176 NCT01941537 Potentially interesting in SS/CTCL NCT02340234 NCT00676884 NCT02542124 X In combination with TSEB NCT02277769 NCT00611208 X Not likely to be used in AD NCT number (https://clinicaltrials.gov/) NCT01728805 X Current study in comparison to HDAC inhibitor 1 3 3 2 2 2 2 2 2 2 na na na na na na na na Phase Table 1: Current and potential immunologic therapeutic targets in S Nivolumab Pembrolizumab (MK-3475) Tralokinumab Lebrikizumab (TNX-650) Pitrakinra Dupilumab (REGN668) A-dmDT390-bisFv(UCHT1) Immunotoxin Substance ∗ ´ ezary syndrome, CTCL: cutaneous T cell lymphoma, AD: atopic dermatitis, HDAC: histone deacetylase, na: not active in SS/CTCL or AD, NHL: non-Hodgkin IL-18 SB-485232 PD-L1 Avelumab PD-1 IL-22 Fezakinumab (ILV-094) TLR7 Imiquimod TLR9 CPG 7909 IL-13 IL-12 NM-IL-12 CD52 Alemtuzumab CTLA4 Ipilimumab CXCR4 Plerixafor IL-4R CD25 Denileukin diftitox CD3 Target CCR4 Mogamulizumab (KW-0761) TSEB: total skin electron beam, and HL: Hodgkin’s lymphoma. SS: S 8 BioMed Research International being tested in solid tumors and systemic lymphoma. As mentioned above the synergism between IL-4 and IL- Though not yet studied in CTCL, they probably will be in 13seemstocontributetotumorcellgrowth.Withregardto near future. this, targeting IL-4 or IL-4-receptor, respectively, would be CCR4 is the receptor for CCL17 and CCL22 and is another therapeutic option. A human monoclonal antibody expressed mainly on CD4+ T cells with Th2 polarization. It against the IL-4-receptor (IL4R), dupilumab,revolutionizes can also be detected on other cells of the immune system like the therapy in AD [140–142]. It is directed against the shared macrophages and dendritic and NK cells [129]. In SS CCR4 alpha subunit of the IL4R and, by IL4R blockage, it modulates expression could be observed in the peripheral blood and in signaling of both the IL-4 and IL-13 pathway. Exploration of theskin[77]. IL-4 in SS showed a significant elevation of IL-4 positive cells Mogamulizumab (Poteligeo5) (Anti-CC chemokine recep- compared to inflammatory dermatoses [143]. This finding tor4(CCR4))is a new humanized monoclonal antibody would support the rationale for treating CTCL patients, obtainable in clinical studies for use in CTCL and SS and especially SS patients, with IL-13/IL-4 inhibitors. is currently in a phase 3 study in comparison with the Resimmune5 (or A-dmDT390-bisFv(UCHT1) Immuno- HDAC inhibitor vorinostat (Zolinza5). In a phase 1/2 study toxin) (Angimmune LLC) is a recombinant immunotoxin in patients with CTCL, the subgroup of 19 SS patients selectively targeting the CD3 receptor and temporarily showed the highest overall response rates with 47.1% [130]. depleting all T cells. It has been shown that sensitivity of Mogamulizumab leads to a depletion of CCR4+ malignant T malignant T cells to this drug is 30 times higher compared to cells and CCR4+Tregs. This mechanism is of great potential normal resting T cells. The drug may have an immunomod- in T cell lymphoma therapy [131]. As mentioned above T ulatory effect by activating novel naive T cells contributing cells are CCR4+ in both AD and SS [132]. Whether the to further deletion of residual tumor cells [144]. Resimmune depletion of Tregs via the CCR4 receptor might be beneficial has been investigated in a clinical phase 2 study in patients in atopic patients has yet to be shown. A phase 1 trial has been with CTCL, including patients with SS. The results are not conducted in patients with asthma. The results are still being yet published. evaluated. Ten different types of toll-like-receptors (TLR) have been described in human until today. Their localization on 5.3. Immunologic Targets in the Environment. Interleukin 12 antigen-presenting plasmacytoid and myeloid dendritic cells (IL-12) is a promising target in CTCL and inflammatory and their role in the immune response in cancer render them skin diseases. In pediatric AD patients IL-12 levels in serum attractive targets for treatment. Apoptosis of tumor cells (e.g., have been reported to be high [133]. Interestingly, the IFN under radiotherapy) and treatment with TLR may lead to syn- serumlevels,whichwouldbeexpectedtoincreaseby ergistic effects [145, 146]. TLR7 agonist (imiquimod/Aldara5) stimulation through IL-12, are comparably low, leading to is available for topical application and is approved in Europe the assumption that the normal IL-12/IFN pathway is not for treatment of superficial basal cell carcinoma and actinic intact [134]. Reduced IL-12R beta (2) mRNA expression keratoses.ItseffectshavealsobeenshownforCTCLpatients may be the cause of low IL-12 receptor expression. High with response rates up to 50% [147–149]. TLR9 agonist (CPG IL-12 without a binding receptor would have no cellular 7909) has been applied subcutaneously to SS and MF in effects and would explain the low IFN serum levels [135]. a phase I study enrolling 28 patients. Clinical response IL-12 substitution would in this case not be beneficial for ratewas32%[150].Inanotherstudy,TLR9agonistwas AD patients. In CTCL induction of lesion regression and injected intralesionally and combined with radiation in 14 cytotoxic T cell responses have been described under IL-12- MF patients leading to the immunological effect of an in therapy [136]. A phase 2 study with NM-IL-12 (recombinant situ vaccination. The overall response rate was 35.7% [151]. human IL-12) should start in November 2015 as a single arm, In mouse the TLR9 agonist was combined with ibrutinib, open-label, nonrandomized study with NM-IL-12 (150 ng/kg) which is a Bruton-tyrosine kinase-inhibitor. This approach dosed in combination with low dose total skin electron beam ledtoanenhancementoftheantitumorresponse[152]. (TSEB) in CTCL patients including patients with SS. IL-12 This effect of the combination will have to be evaluated in this setting acts as immunotherapy to increase antitumor in humans and maybe become a therapy option in the efficacy against CTCL, supposedly reducing skin-related future. toxicity. Anti-CXCR4 is a potential treatment option for patients IL-13 has just recently been described as a contributor with SS. Because of CXCR4 overexpression in SS and MF, to growth of tumor cells in CTCL. IL-4 and IL-13 seem to clinical studies for this target would be interesting [84]. act synergistically in this setting [137, 138]. Based on this For systemic lymphoproliferative diseases like acute myeloid detection, blockage of IL-13 receptor might be beneficial. An leukemia the antibody plerixafor (Mozobil5) is in use. Though IL-13-R-inhibitor is available in experimental studies under there is not yet a study running in CTCL, exploration of this the name lebrikizumab (TNX-650).Thedrughasbeenused therapy in near future is very likely. in a phase I study in Hodgkin’s lymphoma and is currently IL-22 is produced by T cells. Primarily its role in inflam- under investigation in a phase 2 study in AD. Results for these mation in human skin has been in focus, for example, in two studies are not yet published. For asthmatic patients the AD,butobviouslyitisalsoinvolvedinthepathogenesis application of lebrikizumab showed improvement and was of malignant skin proliferation [153]. In CTCL elevated considered as safe and of good tolerability [139]. Another IL- levels of mRNA and protein levels were detected for IL- 13-R-inhibitor that is in use in clinical studies is tralokinumab. 22 [154]. Fezakinumab (ILV-094), a monoclonal antibody BioMed Research International 9 against IL-22, is being investigated in a phase 2 study in AD. mechanisms orchestrated by malignant T cells as well as Experimental use in CTCL could give interesting results. enhancing the local immune response could be beneficial. Another interleukin that could be important in signaling Further exploration is necessary to prolong the life of patients in CTCL is IL-18.IL-18hasbeeninterrelatedtolinking and improve their quality of life. inflammatory immune responses and tumor progression [155]. In skin lesions increased IL-18 expression has been Competing Interests detected and potentially contributing to the elevation to serumlevelsinpatientswithCTCLandcutaneousNKcell The authors have no conflict of interests to declare. lymphoma [156, 157]. High levels of IL-18 have been described in other skin malignancies as well. This might therefore be Acknowledgments either a shared mechanism in skin tumorigenesis or part of an immune response mechanism [158]. Assuming that high lev- This work was supported by the Swiss National Science els of IL-18 are beneficial in cancer or lymphoma, respectively, Foundation (PMPDP3 151326) and the Forschungskredit of human recombinant IL-18 (SB-485232) was administered in theUniversityofZurich(FK-14-032). combination with rituximab in a phase 1 study to patients with non-Hodgkin’s lymphoma (NHL) revealing a response rate of 26% [159]. Phase 2 studies in NHL are ongoing. Further References workup is needed in patients with SS to clarify the role of [1]J.J.Campbell,R.A.Clark,R.Watanabe,andT.S.Kupper, IL-18 and to be able to evaluate if patients would profit from “Sezary´ syndrome and mycosis fungoides arise from distinct this treatment. In AD serum levels for IL-18 are found to be T-cell subsets: a biologic rationale for their distinct clinical elevated likewise [160, 161]. Whether IL-18 application would behaviors,” Blood,vol.116,no.5,pp.767–771,2010. be valuable for the patient is discussed controversially. Some [2] R. A. Clark, J. B. Shackelton, R. Watanabe et al., “High-scatter authors propose that IL-18 might trigger the development of T cells: a reliable biomarker for malignant T cells in cutaneous skin lesions in AD [162]. T-cell lymphoma,” Blood,vol.117,no.6,pp.1966–1976,2011. [3]R.Willemze,E.S.Jaffe,G.Burgetal.,“WHO-EORTCclassi- 5.4. Others. In addition to all of the above-mentioned treat- fication for cutaneous lymphomas,” Blood,vol.105,no.10,pp. ments in CTCL, there are many more that may have, not 3768–3785, 2005. directly but indirectly, effects on the immune response. Epi- [4]M.H.Imam,P.J.Shenoy,C.R.Flowers,A.Phillips,andM.J. genetic modifiers like histone deacetylase (HDAC) inhibitors Lechowicz, “Incidence and survival patterns of cutaneous T-cell or substances that interact with JAK/Stat or NF𝜅Bmay lymphomas in the United States,” Leukemia and Lymphoma,vol. have influence on disease evolution. Furthermore, there 54,no.4,pp.752–759,2013. are proteasome inhibitors like bortezomib (VELCADE5) [5]P.T.Bradford,S.S.Devesa,W.F.Anderson,andJ.R.Toro, that have been used in SS and targeted therapy directed “Cutaneous lymphoma incidence patterns in the United States: a population-based study of 3884 cases,” Blood, vol. 113, no. 21, against molecules expressed on cancer cells and linked to pp. 5064–5073, 2009. cytotoxic agents like the already well-established anti-CD30 5 [6] J. J. Scarisbrick, H. M. Prince, M. H. Vermeer et al., “Cutaneous antibody-drug conjugate brentuximab vedotin (Adcetris ) Lymphoma International Consortium study of outcome in [163]. Expression of CD30 is not exclusively restricted to advanced stages of mycosis fungoides and Sezary´ syndrome: cancer cells, though there are only few cell types in a healthy effect of specific prognostic markers on survival and develop- individual with CD30 surface markers [164]. Tyrosine kinase mentofaprognosticmodel,”Journal of Clinical Oncology,vol. inhibitors like dasatinib have been developed. B-lymphoid 33,no.32,pp.3766–3773,2015. tyrosine kinase is a designated oncogene that could be [7]E.Olsen,E.Vonderheid,N.Pimpinellietal.,“Revisionsto targeted by dasatinib in CTCL [165]. A study has been the staging and classification of mycosis fungoides and Sezary conducted in metastatic and/or not surgically removable syndrome: a proposal of the International Society for Cutaneous lesions in various types of lymphoma, including CTCL and Lymphomas (ISCL) and the cutaneous lymphoma task force SS. Results have not yet been published. of the European Organization of Research and Treatment of Cancer (EORTC),” Blood,vol.110,no.6,pp.1713–1722,2007. 6. Conclusion [8]E.C.Vonderheid,M.G.Bernengo,G.Burgetal.,“Update on erythrodermic cutaneous T-cell lymphoma: report of the Overlapping features in immune responses in both diseases international society for cutaneous lymphomas,” Journal of the might be used for transferring knowledge for target molecules American Academy of Dermatology,vol.46,no.1,pp.95–106, 2002. from one entity to the other. [9]C.D.Klemke,N.Booken,C.Weissetal.,“Histopathological A better understanding of the underlying mechanisms and immunophenotypical criteria for the diagnosis of Sezary´ and new markers will hopefully yield further improvement syndrome in differentiation from other erythrodermic skin of targeted treatment options in both AD and SS. Several diseases: a European Organisation for Research and Treatment promising diagnostic and clinical markers and new treatment of Cancer (EORTC) Cutaneous Lymphoma Task Force Study of options targeting checkpoints in the immune system have 97 cases,” British Journal of Dermatology,vol.173,no.1,pp.93– already been discovered and are implemented in clinical 105, 2015. studies both for SS and for AD. In SS, there is still a need [10] S. Weidinger and N. Novak, “Atopic dermatitis,” The Lancet,vol. for efficacious therapy. Blocking the immunosuppressive 387, no. 10023, pp. 1109–1122, 2016. 10 BioMed Research International

[11] J. S. Margolis, K. Abuabara, W. Bilker, O. Hoffstad, and D. J. treated with long-term corticosteroids for asthma and atopic- Margolis, “Persistence of mild to moderate atopic dermatitis,” like dermatitis,” Journal of Dermatologic Surgery and Oncology, JAMA Dermatology,vol.150,no.6,pp.593–600,2014. vol. 12, no. 10, pp. 1089–1093, 1986. [12] J. I. Silverberg and J. M. Hanifin, “Adult eczema prevalence and [27] E. C. Vonderheid, “On the diagnosis of erythrodermic cuta- associations with asthma and other health and demographic neous T-cell lymphoma,” Journal of Cutaneous Pathology,vol. factors: a US population-based study,” Journal of Allergy and 33, supplement 1, pp. 27–42, 2006. Clinical Immunology, vol. 132, no. 5, pp. 1132–1138, 2013. [28] R. Russell-Jones, “Diagnosing erythrodermic cutaneous T-cell [13] T. Biedermann, Y. Skabytska, S. Kaesler, and T. Volz, “Regula- lymphoma,” British Journal of Dermatology,vol.153,no.1,pp. tion of T cell immunity in atopic dermatitis by microbes: the Yin 1–5, 2005. and Yang of cutaneous inflammation,” Frontiers in Immunology, [29] P. E. LeBoit, “Simulators of cutaneous lymphoma where should vol. 6, article 353, 2015. our efforts go?” American Journal of Clinical Pathology,vol.139, [14] H. Suga, M. Sugaya, T. Miyagaki et al., “Skin barrier dysfunction no. 4, pp. 414–417, 2013. and low antimicrobial peptide expression in cutaneous t-cell [30] T. Bieber, “Atopic dermatitis,” New England Journal of Medicine, lymphoma,” Clinical Cancer Research,vol.20,no.16,pp.4339– vol. 358, no. 14, pp. 1430–1494, 2008. 4348, 2014. [31] A. Katsarou and M. C. Armenaka, “Atopic dermatitis in older [15] S. Kezic, N. Novak, I. Jakasa et al., “Skin barrier in atopic patients: particular points,” JournaloftheEuropeanAcademyof dermatitis,” Frontiers in Bioscience,vol.19,no.3,pp.542–556, Dermatology and Venereology,vol.25,no.1,pp.12–18,2011. 2014. [32] E. Ozkaya, “Adult-onset atopic dermatitis,” Journal of the Amer- [16] R. Talpur, R. Bassett, and M. Duvic, “Prevalence and treatment ican Academy of Dermatology,vol.52,no.4,pp.579–582,2005. of Staphylococcus aureus colonization in patients with mycosis [33] E. Guenova, R. Watanabe, J. E. Teague et al., “TH2 cytokines fungoides and Sezary´ syndrome,” British Journal of Dermatol- from malignant cells suppress TH1 responses and enforce a ogy,vol.159,no.1,pp.105–112,2008. global TH2 bias in leukemic cutaneous T-cell lymphoma,” [17] E. Guenova, W. Hoetzenecker, S. Rozati, M. P. Levesque, R. Clinical Cancer Research,vol.19,no.14,pp.3755–3763,2013. Dummer, and A. Cozzio, “Novel therapies for cutaneous T- [34] E. Guenova, T. Volz, K. Sauer et al., “IL-4-mediated fine tuning cell lymphoma: what does the future hold?” Expert Opinion on of IL-12p70 production by human DC,” European Journal of Investigational Drugs,vol.23,no.4,pp.457–467,2014. Immunology,vol.38,no.11,pp.3138–3149,2008. [35] M. H. Vermeer, R. van Doorn, D. Dukers, M. W. Bekkenk, C. [18] T. Miyagaki and M. Sugaya, “Erythrodermic cutaneous T-cell + lymphoma: how to differentiate this rare disease from atopic J. L. M. Meijer, and R. Willemze, “CD8 T cells in cutaneous T- dermatitis,” Journal of Dermatological Science,vol.64,no.1,pp. cell lymphoma: expression of cytotoxic proteins, Fas ligand, and 1–6, 2011. killing inhibitory receptors and their relationship with clinical behavior,” Journal of Clinical Oncology,vol.19,no.23,pp.4322– [19] J. C. Cather, E. A. Vance, and M. A. Menter, “Diverse cutaneous 4329, 2001. manifestations associated with a single disease,” Proceedings (Baylor University. Medical Center),vol.15,no.4,pp.433–436, [36] G. Goteri, A. Filosa, B. Mannello et al., “Density of neoplastic 2002. lymphoid infiltrate, CD8+ T cells, and CD1a+ dendritic cells in mycosis fungoides,” Journal of Clinical Pathology,vol.56,no.6, [20] L. Parmentier, C. Durr,¨ E. Vassella, H. Beltraminelli, L. Bor- pp. 453–458, 2003. radori, and E. Haneke, “Specific nail alterations in cutaneous T-cell lymphoma: successful treatment with topical [37] S. I. Jawed, P. L. Myskowski, S. Horwitz, A. Moskowitz, and mechlorethamine,” Archives of Dermatology, vol. 146, no. 11, pp. C. Querfeld, “Primary cutaneous T-cell lymphoma (mycosis 1287–1291, 2010. fungoides and Sezary´ syndrome): part I. Diagnosis: clinical and histopathologic features and new molecular and biologic [21]M.Y.Bi,J.L.Curry,A.M.Christianoetal.,“Thespectrum markers,” Journal of the American Academy of Dermatology,vol. of hair loss in patients with mycosis fungoides and Sezary´ 70, no. 2, pp. 205.e1–205.e16, 2014. syndrome,” Journal of the American Academy of Dermatology, [38] L. M. Gjerdrum, A. Woetmann, N. Odum et al., “FOXP3+ vol.64,no.1,pp.53–63,2011. regulatory T cells in cutaneous T-cell lymphomas: association [22] B.E.Jr.Cook,G.B.Bartley,M.R.Pittelkow,andR.H.Kennedy, with disease stage and survival,” Leukemia,vol.21,no.12,pp. “Ophthalmic abnormalities in patients with cutaneous T-cell 2512–2518, 2007. lymphoma,” Transactions of the American Ophthalmological [39] T. Krejsgaard, N. Odum, C. Geisler, M. A. Wasik, and A. Society, vol. 96, pp. 309–327, 1998. Woetmann, “Regulatory T cells and immunodeficiency in [23] K. Mehrany, R. A. El-Azhary, S. A. Bouwhuis, and M. R. mycosis fungoides and Sezary´ syndrome,” Leukemia,vol.26,no. Pittelkow, “Cutaneous T-cell lymphoma and atopy: is there an 3,pp.424–432,2012. association?” British Journal of Dermatology,vol.149,no.5,pp. [40]X.Ni,P.Hazarika,C.Zhang,R.Talpur,andM.Duvic,“Fas 1013–1017, 2003. ligand expression by neoplastic T lymphocytes mediates elimi- + [24] G. Rajka and R. K. Winkelmann, “Atopic dermatitis and Sezary nation of CD8 cytotoxic T lymphocytes in mycosis fungoides: syndrome,” Archives of Dermatology,vol.120,no.1,pp.83–84, a potential mechanism of tumor immune escape?” Clinical 1984. Cancer Research,vol.7,no.9,pp.2682–2692,2001. [25]C.W.VanHaselen,J.Toonstra,A.H.Preesman,S.C.J.Van [41] C. L. Berger, R. Tigelaar, J. Cohen et al., “Cutaneous T- DerPutte,C.A.F.M.Bruijnzeel-Koomen,andW.A.Van cell lymphoma: malignant proliferation of T-regulatory cells,” Vloten, “Sezary syndrome in a young man with severe atopic Blood,vol.105,no.4,pp.1640–1647,2005. dermatitis,” British Journal of Dermatology,vol.140,no.4,pp. [42]R.Dummer,P.W.Heald,F.O.Nestleetal.,“Sezary´ syndrome 704–707, 1999. T-cell clones display T-helper 2 cytokines and express the [26] E. A. Abel, B. J. Nickoloff, D. M. Shelby, W. Watson, and accessory factor-1 (interferon-gamma receptor beta-chain),” G. S. Wood, “Tumor stage mycosis fungoides in a patient Blood,vol.88,no.4,pp.1383–1389,1996. BioMed Research International 11

[43]T.Krejsgaard,U.Ralfkiaer,E.Clasen-Lindeetal.,“Malignant [59] E. J. Kim, S. Hess, S. K. Richardson et al., “Immunopathogenesis cutaneous T-cell lymphoma cells express IL-17 utilizing the and therapy of cutaneous T cell lymphoma,” The Journal of Jak3/stat3 signaling pathway,” Journal of Investigative Dermatol- Clinical Investigation, vol. 115, no. 4, pp. 798–812, 2005. ogy,vol.131,no.6,pp.1331–1338,2011. [60]D.Lu,M.Duvic,L.J.Medeiros,R.Luthra,D.M.Dorfman,and [44] E. Papadavid, J. Economidou, A. Psarra et al., “The relevance D. Jones, “The T-cell chemokine receptor CXCR3 is expressed of peripheral blood T-helper 1 and 2 cytokine pattern in highly in low-grade mycosis fungoides,” American Journal of the evaluation of patients with mycosis fungoides and Sezary´ Clinical Pathology, vol. 115, no. 3, pp. 413–421, 2001. syndrome,” British Journal of Dermatology,vol.148,no.4,pp. [61] M. Grewe, S. Walther, K. Gyufko, W. Czech, E. Schopf, and J. 709–718, 2003. Krutmann, “Analysis of the cytokine pattern expressed in situ [45] B. R. Vowels, M. Cassin, E. C. Vonderheid, and A. H. Rook, in inhalant allergen patch test reactions of atopic dermatitis “Aberrant cytokine production by sezary syndrome patients: patients,” Journal of Investigative Dermatology,vol.105,no.3, cytokine secretion pattern resembles murine Th2 cells,” Journal pp.407–410,1995. of Investigative Dermatology,vol.99,no.1,pp.90–94,1992. [62] R. A. Taha, D. Y. M. Leung, O. Ghaffar, M. Boguniewicz, and [46] L. Kari, A. Loboda, M. Nebozhyn et al., “Classification and Q. Hamid, “In vivo expression of cytokine receptor mRNA in prediction of survival in patients with the leukemic phase of atopic dermatitis,” Journal of Allergy and Clinical Immunology, cutaneousTcelllymphoma,”Journal of Experimental Medicine, vol. 102, no. 2, pp. 245–250, 1998. vol. 197, no. 11, pp. 1477–1488, 2003. [63] S. Kaesler, T. Volz, Y. Skabytska et al., “Toll-like receptor [47] B. R. Vowels, S. R. Lessin, M. Cassin et al., “Th2 cytokine mRNA 2 ligands promote chronic atopic dermatitis through IL-4- expression in skin in cutaneous T-cell lymphoma,” Journal of mediated suppression of IL-10,” Journal of Allergy and Clinical Investigative Dermatology,vol.103,no.5,pp.669–673,1994. Immunology,vol.134,no.1,pp.92.e6–99.e6,2014. [48] A. Rubben,¨ W. Kempf, M. E. Kadin, D. R. Zimmermann, [64]M.Kasprzycka,Q.Zhang,A.Witkiewiczetal.,“𝛾c-signaling and G. Burg, “Multilineage progression of genetically unstable cytokines induce a regulatory T cell phenotype in malignant + tumor subclones in cutaneous T-cell lymphoma,” Experimental CD4 Tlymphocytes,”Journal of Immunology,vol.181,no.4, Dermatology,vol.13,no.8,pp.472–483,2004. pp. 2506–2512, 2008. [49] N. Yawalkar, K. Ferenczi, D. A. Jones et al., “Profound loss [65] T. Krejsgaard, L. M. Gjerdrum, E. Ralfkiaer et al., “Malignant of T-cell receptor repertoire complexity in cutaneous T-cell Tregs express low molecular splice forms of FOXP3 in Sezary´ lymphoma,” Blood,vol.102,no.12,pp.4059–4066,2003. syndrome,” Leukemia,vol.22,no.12,pp.2230–2239,2008. [50] D. Y. Leung and T. Bieber, “Atopic dermatitis,” The Lancet,vol. [66] M. Bagot, M. Nikolova, F. Schirm-Chabanette, J. Wechsler, 361,no.9352,pp.151–160,2003. L. Boumsell, and A. Bensussan, “Crosstalk between tumor T [51] J. K. Gittler, A. Shemer, M. Suarez-Farinas et al., “Progressive lymphocytes and reactive T lymphocytes in cutaneous T cell activation of T(H)2/T(H)22 cytokines and selective epidermal lymphomas,” Annals of the New York Academy of Sciences,vol. proteins characterizes acute and chronic atopic dermatitis,” 941, pp. 31–38, 2001. JournalofAllergyandClinicalImmunology,vol.130,no.6,pp. 1344–1354, 2012. [67] H. T. Jin, R. Ahmed, and T. Okazaki, “Role of PD-1 in regulating T-cell immunity,”in Negative Co-Receptors and Ligands,vol.350 [52] T. Biedermann, M. Rocken,andJ.M.Carballido,“TH1and¨ of Current Topics in Microbiology and Immunology,pp.17–37, TH2 lymphocyte development and regulation of TH cell- Springer, Berlin, Germany, 2011. mediated immune responses of the skin,” JournalofInvestigative Dermatology Symposium Proceedings,vol.9,no.1,pp.5–14, [68] F. C¸etinozman,¨ P. M. Jansen, M. H. Vermeer, and R. Willemze, 2004. “Differential expression of programmed death-1 (PD-1) in [53] T. Volz, Y. Skabytska, E. Guenova et al., “Nonpathogenic Sezary´ syndrome and mycosis fungoides,” Archives of Derma- bacteria alleviating atopic dermatitis inflammation induce IL- tology,vol.148,no.12,pp.1379–1385,2012. 10-producing dendritic cells and regulatory Tr1 cells,” Journal of [69] S. Samimi, B. Benoit, K. Evans et al., “Increased programmed Investigative Dermatology,vol.134,no.1,pp.96–104,2014. death-1 expression on CD4+ T cells in cutaneous T-cell lym- [54] Y. Skabytska, F. Wolbing,¨ C. Gunther¨ et al., “Cutaneous innate phoma: implications for immune suppression,” Archives of immune sensing of toll-like receptor 2-6 ligands suppresses T Dermatology,vol.146,no.12,pp.1382–1388,2010. cell immunity by inducing myeloid-derived suppressor cells,” [70] E. Contassot, K. Kerl, S. Roques et al., “Resistance to FasL Immunity,vol.41,no.5,pp.762–775,2014. and tumor necrosis factor-related apoptosis-inducing ligand- [55]V.Soumelis,P.A.Reche,H.Kanzleretal.,“Humanepithelial mediated apoptosis in Sezary´ syndrome T-cells associated cells trigger dendritic cell−mediated allergic inflammation by with impaired death receptor and FLICE-inhibitory protein producing TSLP,” Nature Immunology,vol.3,no.7,pp.673–680, expression,” Blood, vol. 111, no. 9, pp. 4780–4787, 2008. 2002. [71] O. Dereure, E. Levi, E. C. Vonderheid, and M. E. Kadin, [56] M. Matsumoto, C. Ra, K. Kawamoto et al., “IgE hyperpro- “Infrequent Fas mutations but no Bax or p53 mutations in early duction through enhanced tyrosine phosphorylation of janus mycosis fungoides: a possible mechanism for the accumulation kinase 3 in NC/Nga mice, a model for human atopic dermatitis,” of malignant T lymphocytes in the skin,” Journal of Investigative Journal of Immunology,vol.162,no.2,pp.1056–1063,1999. Dermatology,vol.118,no.6,pp.949–956,2002. [57] Y. Kataoka, “Thymus and activation-regulated chemokine as a [72]O.Dereure,P.Portales,J.Clot,andJ.-J.Guilhou,“Decreased + clinical biomarker in atopic dermatitis,” Journal of Dermatology, expression of Fas (APO-1/CD95) on peripheral blood CD4 T vol. 41, no. 3, pp. 221–229, 2014. lymphocytes in cutaneous T-cell lymphomas,” British Journal of [58] J. Thijs, T. Krastev, S. Weidinger et al., “Biomarkers for atopic Dermatology,vol.143,no.6,pp.1205–1210,2000. dermatitis: a systematic review and meta-analysis,” Current [73] J. Wu, M. Nihal, J. Siddiqui, E. C. Vonderheid, and G. S. Opinion in Allergy and Clinical Immunology,vol.15,no.5,pp. Wood,“LowFAS/CD95expressionbyCTCLcorrelateswith 453–460, 2015. reduced sensitivity to apoptosis that can be restored by FAS 12 BioMed Research International

upregulation,” Journal of Investigative Dermatology,vol.129,no. [87] M. Sugaya, “Chemokines and cutaneous lymphoma,” Journal of 5, pp. 1165–1173, 2009. Dermatological Science,vol.59,no.2,pp.81–85,2010. [74] C. L. Jones, E. M. Wain, C.-C. Chu et al., “Downregulation of fas [88] U. Forssmann, M. Uguccioni, P. Loetscher et al., “Eotaxin- gene expression in sezary´ syndrome is associated with promoter 2, a novel CC chemokine that is selective for the chemokine hypermethylation,” JournalofInvestigativeDermatology,vol. receptor CCR3, and acts like eotaxin on human eosinophil and 130, no. 4, pp. 1116–1125, 2010. basophil leukocytes,” Journal of Experimental Medicine,vol.185, [75]R.vanDoorn,R.Dijkman,M.H.Vermeer,T.M.Starink, no.12,pp.2171–2176,1997. R. Willemze, and C. P. Tensen, “A novel splice variant of the [89] F. Sallusto, C. R. Mackay, and A. Lanzavecchia, “Selective FasgeneinpatientswithcutaneousT-celllymphoma,”Cancer expression of the eotaxin receptor CCR3 by human T helper 2 Research,vol.62,no.19,pp.5389–5392,2002. cells,” Science,vol.277,no.5334,pp.2005–2007,1997. [76] Y. Reiss, A. E. Proudfoot, C. A. Power, J. J. Campbell, and E. [90] T. Miyagaki, M. Sugaya, H. Fujita et al., “Eotaxins and CCR3 C. Butcher, “CC chemokine receptor (CCR)4 and the CCR10 interaction regulates the Th2 environment of cutaneous T-Cell ligand cutaneous T cell-attracting chemokine (CTACK) in lym- lymphoma,” Journal of Investigative Dermatology,vol.130,no.9, phocyte trafficking to inflamed skin,” Journal of Experimental pp. 2304–2311, 2010. Medicine,vol.194,no.10,pp.1541–1547,2001. [91] N.Berkman,S.Ohnona,F.K.Chung,andR.Breuer,“Eotaxin-3 [77]K.Ferenczi,R.C.Fuhlbrigge,J.L.Pinkus,G.S.Pinkus,and but not eotaxin gene expression is upregulated in asthmatics 24 T. S. Kupper, “Increased CCR4 expression in cutaneous T cell hoursafterallergenchallenge,”American Journal of Respiratory lymphoma,” Journal of Investigative Dermatology,vol.119,no.6, Cell and Molecular Biology,vol.24,no.6,pp.682–687,2001. pp. 1405–1410, 2002. [92] S. Kagami, T. Kakinuma, H. Saeki et al., “Significant elevation of [78] T. Kakinuma, M. Sugaya, K. Nakamura et al., “Thymus and serum levels of eotaxin-3/CCL26, but not of eotaxin-2/CCL24, activation-regulated chemokine (TARC/CCL17) in mycosis in patients with atopic dermatitis: serum eotaxin-3/CCL26 fungoides: serum TARC levels reflect the disease activity of levels reflect the disease activity of atopic dermatitis,” Clinical mycosis fungoides,” Journal of the American Academy of Der- and Experimental Immunology,vol.134,no.2,pp.309–313, matology,vol.48,no.1,pp.23–30,2003. 2003. [79] C. Robert and T. S. Kupper, “Inflammatory skin diseases, T cells, [93] S. Ebner, V. A. Nguyen, M. Forstner et al., “Thymic stromal and immune surveillance,” New England Journal of Medicine, lymphopoietin converts human epidermal Langerhans cells vol.341,no.24,pp.1817–1828,1999. into antigen-presenting cells that induce proallergic T cells,” [80]H.Yagi,N.Seo,A.Ohshimaetal.,“Chemokinereceptor Journal of Allergy and Clinical Immunology,vol.119,no.4,pp. expression in cutaneous T cell and NK/T-cell lymphomas: 982–990, 2007. Immunohistochemical staining and in vitro chemotactic assay,” [94] J. Yoo, M. Omori, D. Gyarmati et al., “Spontaneous atopic American Journal of Surgical Pathology, vol. 30, no. 9, pp. 1111– dermatitis in mice expressing an inducible thymic stromal 1119, 2006. lymphopoietin transgene specifically in the skin,” Journal of Experimental Medicine,vol.202,no.4,pp.541–549,2005. [81] T. Yamaguchi, K. Ohshima, T. Tsuchiya et al., “The com- parisonofexpressionofcutaneouslymphocyte-associated [95] T. Miyagaki and M. Sugaya, “Immunological milieu in mycosis antigen (CLA), and Th1- and Th2-associated antigens in fungoides and Sezary´ syndrome,” JournalofDermatology,vol. mycosis fungoides and cutaneous lesions of adult T-cell 41,no.1,pp.11–18,2014. leukemia/lymphoma,” European Journal of Dermatology,vol.13, [96] N. K. Mollanazar, P. K. Smith, and G. Yosipovitch, “Mediators no. 6, pp. 553–559, 2003. of chronic pruritus in atopic dermatitis: getting the itch out?” [82] M. Sokolowska-Wojdylo, J. Wenzel, E. Gaffal et al., “Circulating Clinical Reviews in Allergy and Immunology,2015. clonal CLA+ and CD4+ T cells in Sezary´ syndrome express the [97] M. Sokołowska-Wojdyło, J. Glen,´ M. Zabłotna et al., “Asso- skin-homing chemokine receptors CCR4 and CCR10 as well ciation of distinct IL-31 polymorphisms with pruritus and as the lymph node-homing chemokine receptor CCR7,” British severity of atopic dermatitis,” Journal of the European Academy Journal of Dermatology,vol.152,no.2,pp.258–264,2005. of Dermatology and Venereology,vol.27,no.5,pp.662–664, [83] T. Kallinich, J. M. Muche, S. Qin, W. Sterry, H. Audring, and 2013. R. A. Kroczek, “Chemokine receptor expression on neoplastic [98]K.Hartmann,N.Wagner,A.Rabenhorstetal.,“SerumIL-31 and reactive T cells in the skin at different stages of mycosis levels are increased in a subset of patients with mastocytosis fungoides,” Journal of Investigative Dermatology,vol.121,no.5, and correlate with disease severity in adult patients,” Journal of pp.1045–1052,2003. Allergy and Clinical Immunology,vol.132,no.1,pp.232–e4,2013. [84] M.G.Narducci,E.Scala,A.Bresinetal.,“SkinhomingofSezary´ [99] M. Mobs,S.Gryzik,A.Haidar,D.Humme,M.Beyer,andS.¨ cells involves SDF-1-CXCR4 signaling and down-regulation of Vandersee, “Analysis of the IL-31 pathway in mycosis fungoides CD26/dipeptidylpeptidase IV,” Blood,vol.107,no.3,pp.1108– and sezary´ syndrome,” Archives of Dermatological Research,vol. 1115, 2006. 307, no. 6, pp. 479–485, 2014. [85] E. Capriotti, E. C. Vonderheid, C. J. Thoburn, E. C. Bright, and [100] P. Y. Ong, T. Ohtake, C. Brandt et al., “Endogenous antimicro- A. D. Hess, “Chemokine receptor expression by leukemic T cells bial peptides and skin infections in atopic dermatitis,” The New of cutaneous T-cell lymphoma: clinical and histopathological England Journal of Medicine, vol. 347, no. 15, pp. 1151–1160, 2002. correlations,” Journal of Investigative Dermatology,vol.127,no. [101]I.Nomura,E.Goleva,M.D.Howelletal.,“Cytokinemilieu 12, pp. 2882–2892, 2007. of atopic dermatitis, as compared to psoriasis, skin prevents [86] C. L. Tendler, J. D. Burton, J. Jaffe et al., “Abnormal cytokine induction of innate immune response genes,” The Journal of expression in Sezary´ and adult T-cell leukemia cells correlates Immunology, vol. 171, no. 6, pp. 3262–3269, 2003. with the functional diversity between these T-cell malignan- [102]M.D.Howell,A.Wollenberg,R.L.Galloetal.,“Catheli- cies,” Cancer Research, vol. 54, no. 16, pp. 4430–4435, 1994. cidin deficiency predisposes to eczema herpeticum,” Journal of BioMed Research International 13

Allergy and Clinical Immunology,vol.117,no.4,pp.836–841, [118] R. Sidbury, D. M. Davis, D. E. Cohen et al., “Guidelines 2006. of care for the management of atopic dermatitis: Section 3. [103]P.I.Axelrod,B.Lorber,andE.C.Vonderheid,“Infections Management and treatment with phototherapy and systemic complicating mycosis fungoides and sezary´ syndrome,” The agents,” Journal of the American Academy of Dermatology,vol. Journal of the American Medical Association, vol. 267, no. 10, pp. 71, no. 2, pp. 327–349, 2014. 1354–1358, 1992. [119] E. A. Olsen, A. H. Rook, J. Zic et al., “Sezary´ syndrome: [104]C.M.Jackow,J.C.Cather,V.Hearne,A.T.Asano,J.M.Musser, immunopathogenesis, literature review of therapeutic options, and M. Duvic, “Association of erythrodermic cutaneous T-cell and recommendations for therapy by the United States Cuta- lymphoma, superantigen- positive Staphylococcus aureus,and neous Lymphoma Consortium (USCLC),” Journal of the Amer- oligoclonal T-cell receptor V𝛽 gene expansion,” Blood,vol.89, ican Academy of Dermatology,vol.64,no.2,pp.352–404,2011. no. 1, pp. 32–40, 1997. [120] J.Guitart,S.C.Wickless,Y.Oyamaetal.,“Long-termremission [105] V. Nguyen, R. H. Huggins, T. Lertsburapa et al., “Cutaneous T- after allogeneic hematopoietic stem cell transplantation for cell lymphoma and Staphylococcus aureus colonization,” Journal refractory cutaneous T-cell lymphoma,” Archives of Dermatol- of the American Academy of Dermatology,vol.59,no.6,pp.949– ogy,vol.138,no.10,pp.1359–1365,2002. 952, 2008. [121] C. Querfeld, S. T. Rosen, T. M. Kuzel et al., “Long-term follow- [106]S.J.K.Gilani,M.Gonzalez,I.Hussain,A.Y.Finlay,andG.K. up of patients with early-stage cutaneous T-cell lymphoma Patel, “Staphylococcus aureus re-colonization in atopic dermati- who achieved complete remission with psoralen plus UV-A tis: beyond the skin,” Clinical and Experimental Dermatology, monotherapy,” Archives of Dermatology,vol.141,no.3,pp.305– vol.30,no.1,pp.10–13,2005. 311, 2005. [107] A. Roll, A. Cozzio, B. Fischer, and P. Schmid-Grendelmeier, [122] Y. Mansouri and E. Guttman-Yassky, “Immune pathways in “Microbial colonization and atopic dermatitis,” Current Opinion atopic dermatitis, and definition of biomarkers through broad in Allergy and Clinical Immunology,vol.4,no.5,pp.373–378, and targeted therapeutics,” Journal of Clinical Medicine,vol.4, 2004. no.5,pp.858–873,2015. [108] S. Nobbe, P. Dziunycz, B. Muhleisen¨ et al., “IL-31 expression by [123] J. Lundin, H. Hagberg, R. Repp et al., “Phase 2 study of inflammatory cells is preferentially elevated in atopic dermati- alemtuzumab (anti-CD52 monoclonal antibody) in patients tis,” Acta Dermato-Venereologica,vol.92,no.1,pp.24–28,2012. with advanced mycosis fungoides/Sezary´ syndrome,” Blood,vol. [109]E.Sonkoly,A.Muller,A.I.Lauermaetal.,“IL-31:anewlink 101, no. 11, pp. 4267–4272, 2003. between T cells and pruritus in atopic skin inflammation,” [124] R.A.Clark,R.Watanabe,J.E.Teagueetal.,“Skineffectormem- Journal of Allergy and Clinical Immunology,vol.117,no.2,pp. ory T cells do not recirculate and provide immune protection 411–417, 2006. in alemtuzumab-treated CTCL patients,” Science Translational [110] V. H. Sommer, O. J. Clemmensen, O. Nielsen et al., “In vivo Medicine,vol.4,no.117,ArticleID117ra7,2012. activation of STAT3 in cutaneous T-cell lymphoma. Evidence [125] K. A. Thursky, L. J. Worth, J. F.Seymour, H. Miles Prince, and M. for an antiapoptotic function of STAT3,” Leukemia,vol.18,no. A. Slavin, “Spectrum of infection, risk and recommendations 7, pp. 1288–1295, 2004. for prophylaxis and screening among patients with lymphopro- [111] K. Asadullah, W.-D. Docke,¨ A. Haeuler, W. Sterry, and H.- liferative disorders treated with alemtuzumab,” British Journal D. Volk, “Progression of mycosis fungoides is associated with of Haematology,vol.132,no.1,pp.3–12,2006. increasing cutaneous expression of interleukin-10 mRNA,” Journal of Investigative Dermatology,vol.107,no.6,pp.833–837, [126] H.M.Gibson,A.Mishra,D.V.Chan,T.S.Hake,P.Porcu,andH. 1996. K. Wong, “Impaired proteasome function activates GATA3 in T cells and upregulates CTLA-4: relevance for Sezary´ syndrome,” [112] T. Krejsgaard, A. Willerslev-Olsen, L. M. Lindahl et al., “Staphy- The Journal of Investigative Dermatology,vol.133,no.1,pp.249– lococcal enterotoxins stimulate lymphoma-associated immune 257, 2013. dysregulation,” Blood,vol.124,no.5,pp.761–770,2014. [113] Y. Tokura, H. Yagi, A. Ohshima et al., “Cutaneous colonization [127]H.K.Wong,A.J.Wilson,H.M.Gibsonetal.,“Increased with staphylococci influences the disease activity ofezary S´ expression of CTLA-4 in malignant T cells from patients with syndrome: a potential role for bacterial superantigens ,” British mycosis fungoides - Cutaneous T-cell lymphoma,” Journal of Journal of Dermatology,vol.133,no.1,pp.6–12,1995. Investigative Dermatology,vol.126,no.1,pp.212–219,2006. [114] H. K. Wong, A. Mishra, T.Hake, and P.Porcu, “Evolving insights [128]A.Patnaik,S.P.Kang,D.Rascoetal.,“Phaseistudyofpem- in the pathogenesis and therapy of cutaneous T-cell lymphoma brolizumab (MK-3475; Anti-PD-1 monoclonal antibody) in (mycosis fungoides and sezary syndrome),” British Journal of patients with advanced solid tumors,” Clinical Cancer Research, Haematology,vol.155,no.2,pp.150–166,2011. vol. 21, no. 19, pp. 4286–4293, 2015. [115] E. Guenova, Y. Skabytska, W. Hoetzenecker et al., “IL-4 abro- [129]Y.Chvatchko,A.J.Hoogewerf,A.Meyeretal.,“Akeyrole for CC chemokine receptor 4 in lipopolysaccharide-induced gates TH17 cell-mediated inflammation by selective silencing of IL-23 in antigen-presenting cells,” Proceedings of the National endotoxic shock,” The Journal of Experimental Medicine,vol.191, Academy of Sciences of the United States of America,vol.112,no. no.10,pp.1755–1764,2000. 7,pp.2163–2168,2015. [130] M. Duvic, L. C. Pinter-Brown, F. M. Foss et al., “Phase 1/2 study [116] R. Dummer, S. Rozati, E. Guenova, and A. Cozzio, “Less can of mogamulizumab, a defucosylated anti-CCR4 antibody, in be more: the impact of chemotherapy on cutaneous T-cell previously treated patients with cutaneous T-cell lymphoma,” lymphomas,” Future Oncology,vol.9,no.8,pp.1061–1064,2013. Blood,vol.125,no.12,pp.1883–1889,2015. [117] F. Trautinger, R. Knobler, R. Willemze et al., “EORTC con- [131] X. Ni, T. Langridge, and M. Duvic, “Depletion of regulatory sensus recommendations for the treatment of mycosis fun- T cells by targeting CC chemokine receptor type 4 with goides/Sezary´ syndrome,” European Journal of Cancer,vol.42, mogamulizumab,” OncoImmunology,vol.4,no.7,ArticleID no. 8, pp. 1014–1030, 2006. e1011524, 2015. 14 BioMed Research International

[132] M. Sugaya, S. Morimura, H. Suga et al., “CCR4 is expressed on in mycosis fungoides,” European Journal of Dermatology,vol.18, infiltrating cells in lesional skin of early mycosis fungoides and no. 2, pp. 148–152, 2008. atopic dermatitis,” The Journal of Dermatology,vol.42,no.6,pp. [148] N. Onsun, H. Ufacik, Y.Kural, E. Topc¸u, and A. Somay, “Efficacy 613–615, 2015. of imiquimod in solitary plaques of mycosis fungoides,” Inter- [133] D. Piancatelli, L. Bellotta, T. Del Beato, M. Duse, and M. national Journal of Tissue Reactions,vol.27,no.4,pp.167–172, R. Della Penna, “Total il-12 levels are increased in paediatric 2005. atopic dermatitis: correlations with age and disease severity,” [149]M.J.Deeths,J.T.Chapman,R.P.Dellavalle,C.Zeng,andJ.L. International Journal of Immunopathology and Pharmacology, Aeling, “Treatment of patch and plaque stage mycosis fungoides vol. 21, no. 2, pp. 359–365, 2008. with imiquimod 5% cream,” Journal of the American Academy [134] N. Kondo, E. Matsui, H. Kaneko et al., “Reduced interferon- of Dermatology,vol.52,no.2,pp.275–280,2005. 𝛾 𝛽 production and mutations of the interleukin-12 receptor 2 [150] Y. H. Kim, M. Girardi, M. Duvic et al., “Phase i trial of a Toll- chaingeneinatopicsubjects,”International Archives of Allergy like receptor 9 agonist, PF-3512676 (CPG 7909), in patients with and Immunology,vol.124,no.1–3,pp.117–120,2001. treatment-refractory, cutaneous T-cell lymphoma,” Journal of [135] J.K.Eun,M.L.Won,S.H.Jung,H.R.Nam,S.J.Dong,andR.K. the American Academy of Dermatology,vol.63,no.6,pp.975– Jae, “mRNA expression and RNA editing (2451 C-to-U) of IL-12 983, 2010. receptor 𝛽2 in adult atopic patients,” Journal of Korean Medical [151] Y. H. Kim, D. Gratzinger, C. Harrison et al., “In situ vaccination Science,vol.21,no.6,pp.1070–1074,2006. against mycosis fungoides by intratumoral injection of a TLR9 [136] A. H. Rook, G. S. Wood, E. K. Yoo et al., “Interleukin-12 therapy agonist combined with radiation: a phase 1/2 study,” Blood,vol. of cutaneous T-cell lymphoma induces lesion regression and 119, no. 2, pp. 355–363, 2012. cytotoxic T-cell responses,” Blood,vol.94,no.3,pp.902–908, [152] I.Sagiv-Barfi,H.E.Kohrt,L.Burckhardt,D.K.Czerwinski,and 1999. R. Levy, “Ibrutinib enhances the antitumor immune response [137] M. A. Wasik, “IL-13 as a novel growth factor in CTCL,” Blood, induced by intratumoral injection of a TLR9 ligand in mouse vol. 125, no. 18, pp. 2737–2738, 2015. lymphoma,” Blood,vol.125,no.13,pp.2079–2086,2015. [138] L. J. Geskin, S. Viragova, D. B. Stolz, and P. Fuschiotti, [153] H. Fujita, “The role of IL-22 and Th22 cells in human skin “Interleukin-13 is overexpressed in cutaneous T-cell lymphoma diseases,” JournalofDermatologicalScience,vol.72,no.1,pp. cells and regulates their proliferation,” Blood,vol.125,no.18,pp. 3–8, 2013. 2798–2805, 2015. [154] T. Miyagaki, M. Sugaya, H. Suga et al., “IL-22, but not IL- [139] N. A. Hanania, M. Noonan, J. Corren et al., “Lebrikizumab in 17, Dominant environment in cutaneous T-cell lymphoma,” moderate-to-severe asthma: pooled data from two randomised Clinical Cancer Research,vol.17,no.24,pp.7529–7538,2011. placebo-controlled studies,” Thorax,vol.70,no.8,pp.748–756, [155]M.B.Kuppala,S.B.Syed,S.Bandaru,S.Varre,J.Akka, 2015. and H. P. Mundulru, “Immunotherapeutic approach for better [140] L. A. Beck, D. Thac¸i,J.D.Hamiltonetal.,“Dupilumab management of cancer—role of IL-18,” Asian Pacific Journal of treatment in adults with moderate-to-severe atopic dermatitis,” Cancer Prevention, vol. 13, no. 11, pp. 5353–5361, 2012. New England Journal of Medicine,vol.371,no.2,pp.130–139, [156] Y. Amo, Y. Ohta, Y. Hamada, and K. Katsuoka, “Serum levels of 2014. interleukin-18 are increased in patients with cutaneous T-cell [141]J.D.Hamilton,M.Suarez-Fari´ nas,˜ N. Dhingra et al., lymphoma and cutaneous natural killer-cell lymphoma,” The “Dupilumab improves the molecular signature in skin of British Journal of Dermatology,vol.145,no.4,pp.674–676,2001. patients with moderate-to-severe atopic dermatitis,” Journal of [157] K.-I. Yamanaka, R. Clark, R. Dowgiert et al., “Expression of Allergy and Clinical Immunology,vol.134,no.6,pp.1293–1300, interleukin-18 and caspase-1 in cutaneous T-cell lymphoma,” 2014. Clinical Cancer Research,vol.12,no.2,pp.376–382,2006. [142] A.TsianakasandT.A.Luger,“Theanti-IL-4receptoralphaanti- [158] H. Park, D. Byun, T. S. Kim et al., “Enhanced IL-18 expression body dupilumab: facing a new era in treating atopic dermatitis,” in common skin tumors,” Immunology Letters,vol.79,no.3,pp. ExpertOpiniononBiologicalTherapy,vol.15,no.11,pp.1661– 215–219, 2001. 1664, 2015. [159] M. J. Robertson, J. Kline, H. Struemper et al., “Adose-escalation [143] V. Sigurdsson, J. Toonstra, I. C. Bihari, C. A. F. M. Bruijnzeel- study of recombinant human interleukin-18 in combination Koomen, W. A. van Vloten, and T. Thepen, “Interleukin 4 with rituximab in patients with non-hodgkin Lymphoma,” and interferon-𝛾 expression of the dermal infiltrate in patients Journal of Immunotherapy,vol.36,no.6,pp.331–341,2013. with erythroderma and mycosis fungoides. An immuno- [160] M. Aral, O. Arican, M. Gul et al., “The relationship between histochemical study,” Journal of Cutaneous Pathology,vol.27,no. serum levels of total IgE, IL-18, IL-12, IFN-𝛾 and disease severity 9, pp. 429–435, 2000. in children with atopic dermatitis,” Mediators of Inflammation, [144]J.H.Woo,Y.-J.Lee,D.M.Neville,andA.E.Frankel,“Pharma- vol. 2006, Article ID 73098, 4 pages, 2006. cology of anti-CD3 diphtheria immunotoxin in CD3 positive [161]K.Zedan,Z.Rasheed,Y.Farouketal.,“Immunoglobulin T-cell lymphoma trials,” Methods in Molecular Biology,vol.651, E, interleukin-18 and interleukin-12 in patients with atopic pp.157–175,2010. dermatitis: correlation with disease activity,” Journal of Clinical [145] A. H. Rook, “The beauty of TLR agonists for CTCL,” Blood,vol. and Diagnostic Research,vol.9,no.4,pp.WC01–WC05,2015. 119, no. 2, pp. 321–322, 2012. [162]H.Konishi,H.Tsutsui,T.Murakamietal.,“IL-18contributesto [146] A. O. Huen and A. H. Rook, “Toll receptor agonist therapy of the spontaneous development of atopic dermatitis-like inflam- skin cancer and cutaneous T-cell lymphoma,” Current Opinion matory skin lesion independently of IgE/stat6 under specific in Oncology,vol.26,no.2,pp.237–244,2014. pathogen-free conditions,” Proceedings of the National Academy [147] M. C. Mart´ınez-Gonzalez,´ M. M. Verea-Hernando, M. T. Yebra- of Sciences of the United States of America,vol.99,no.17,pp. Pimentel, J. Del Pozo, M. Mazaira, and E. Fonseca, “Imiquimod 11340–11345, 2002. BioMed Research International 15

[163]T.Mehra,K.Ikenberg,R.M.Moosetal.,“Brentuximabasa + treatment for CD30 mycosis fungoides and Sezary´ syndrome,” JAMA Dermatology,vol.151,no.1,pp.73–77,2015. [164]R.Chiarle,A.Podda,G.Prolla,J.Gong,G.J.Thorbecke,and G. Inghirami, “CD30 in normal and neoplastic cells,” Clinical Immunology,vol.90,no.2,pp.157–164,1999. [165] D. L. Petersen, T. Krejsgaard, J. Berthelsen et al., “B-lymphoid tyrosine kinase (Blk) is an oncogene and a potential target for therapy with dasatinib in cutaneous T-cell lymphoma (CTCL),” Leukemia, vol. 28, no. 10, pp. 2109–2112, 2014. Hindawi Publishing Corporation BioMed Research International Volume 2015, Article ID 809406, 10 pages http://dx.doi.org/10.1155/2015/809406

Review Article The Skin as an Early Expression of Malignancies in the Neonatal Age: A Review of the Literature and a Case Series

Vito Mondì, Fiammetta Piersigilli, Guglielmo Salvatori, and Cinzia Auriti

Department of Medical and Surgical Neonatology, Bambino Gesu` Children’s Hospital, Piazza S. Onofrio 4, 00165 Rome, Italy

Correspondence should be addressed to Cinzia Auriti; [email protected]

Received 11 September 2015; Revised 15 November 2015; Accepted 29 November 2015

Academic Editor: Razvigor Darlenski

Copyright © 2015 Vito Mond`ı et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Skin lesions are a frequent finding in childhood, from infancy throughout adolescence. They can arise from many conditions, including infections and inflammation. Most neonatal rashes are benign and self-limiting and require no treatment. Other conditions may be an expression of malignancy or may be a marker for other abnormalities, such as neural tube defects. Therefore, skin lesions require an extensive evaluation and close follow-up to ensure the best possible outcome. This paper briefly reviews the main tumor types presenting with cutaneous involvement in neonates, followed by the description of some patients admitted to our Neonatal Intensive Care Unit with an early skin expression of malignancies.

1. Background births [10]. Leukaemia cutis is reported to be the initial presenting sign in 50% of neonates with leukemia [11, 12]. It Dermatological diseases in neonates are commonly benign is a broad term used to describe any cutaneous localisation of and self-limiting, but they may also herald underlying sys- leukaemic cells in the skin, occurring in 25–64% of patients temic diseases that are potentially life threatening. Neonatal with neonatal acute leukaemia. In approximately 10% of the cancer is a relatively uncommon finding; in UK and USA, the cases, neonates can present with leukaemia cutis, without incidence is approximately one in every 12,500–27,500 live bone marrow involvement, a condition termed aleukaemic births [1], varying from 17 to 121 per million live births in leukaemia cutis [10, 12]. The blueberry muffin rash can be reported series [2, 3]. Neonatal cancer, though uncommon, the initial manifestation both of acute myeloid leukaemia can be already present at birth and a skin lesion can be its first (AML) and of acute lymphoblastic leukaemia (ALL). The sign. Many malignancies can present already at the first exam- pathophysiology underlying the migration of leukaemic cells ination after birth as a “blueberry muffin” rash, characterized intheskinisnotyetunderstoodandthereisnoevidenceofa by localized or generalized reddish-blue papular nodules. specific phenotype associated with cutaneous localization. It In fact, besides congenital infections from the Toxoplasma- has been speculated that the chemokine, integrin, and other Rubella-Cytomegalovirus-Herpes (TORCH) complex and adhesion (ICAM-1) molecules may play a role in skin specific severe haemolysis, many early onset malignancies may cause homing of T and B leukemic cells. CD56 (blast neural cell a similar clinical picture (Table 1 and Table 2) [4–9]. adhesion molecule) expressed on leukemic blasts has been In this paper, we review the most frequent neonatal associated with extramedullary disease in acute leukemia malignancies presenting in the neonate with predominant patients with t(8;21) [13]. Nevertheless, some genetic or skin involvement. chromosomal alterations are found in specific subtypes of leukaemia and have a role in conditioning the severity of the 2. Leukemia clinical course. Rearrangement of the mixed lineage leukaemia (MLL) The incidence of neonatal leukaemia, diagnosed in the first gene, which occurs in the 11q23 translocation, leads to 28 days of life after birth, ranges from 1 to 5 million live aggressive acute leukaemia and may be present in both 2 BioMed Research International

Table 1: Causes of the blueberry muffin rash. papilloedema, retinal vein distension, and retinal haem- orrhage. Respiratory symptoms may include tachypnoea, Congenital TORCH complex dyspnoea, hypoxia with pulmonary infiltrates, or respiratory infections failure [20, 21]. Furthermore, leukemia may manifest itself (i) ABO or Rhesus incompatibility Severe with anemia and bone pain because of extramedullary infil- haemolysis (ii) Hereditary spherocytosis tration. (iii) Twin-twin transfusion (i) Multiple hemangiomas of infancy 2.2. Skin Involvement. Skin involvement is frequently present Congenital (ii) Multifocal lymphangioendotheliomatosis already at birth with multiple randomly distributed subcuta- vascular lesions (iii) Blue rubber bleb nevus syndrome neousnodulesonthetrunkandface,withbluishinfiltrateson (iv) Multiple glomangiomas the whole body surface, similar to blueberry muffin eruption, (i) Leukemia although nodular lesions can be usually small and not bluish, and the infiltrate can be subcutaneous, not nodular (ii) Langerhans cell histiocytosis and with a generalized distribution (Figure 1). The infiltrate (iii) Disseminate neuroblastoma exhibits no blanching response to palpation. In addition Early onset (iv) Rhabdoid tumor to the examination of the peripheral smear to evaluate the malignancies (v) Rhabdomyosarcoma percentage of blast cells, thrombocytopenia or anemia, the (vi) Primitive neuroectodermal tumors bone marrow aspirate to perform cytogenetic analysis, a skin (vii) Choriocarcinoma biopsy is also mandatory. In approximately 10% of cases of congenital leukemia cutis, cutaneous manifestations may be (viii) Myofibromatosis present in the setting of normal bone marrow and peripheral blood smear [22]. The histopathologic analysis of the lesions in leukemia cutis may reveal several different patterns of infiltration. Most commonly they appear as a dense diffuse dermal AML and ALL [11]; it occurs in 34% of this subgroup [14]. infiltrate of pleomorphic leukemic cells, observed in a linear On the other hand, spontaneous remission can occur in a array between collagen bundles in the reticular dermis. subset of neonates affected by AML M4 with t(8;16)(p11;p13). Alternatively, band-like as well as compact nodular infiltrates Nevertheless,recurrencecanoccurinalmosthalfofthecases may be seen. These infiltrates rarely reach the epidermis in the first year of life [15]. and most cases show extension into the subcutis. The nuclei In utero leukaemogenesis has been suspected because ofthesecellsmayrangefrommarkedlyatypicalfoldedor of the finding of high leukemic cell burdens at birth and reniform nuclei to monomorphous and cytological bands. from autopsies of stillborn infants with leukemia and because Determining the leukemic infiltrate subtype almost always of the description of leukemia with identical karyotype requires staining for immunophenotypic markers and their abnormalities in monozygotic twins. An association between association with cryptologic and cytogenetic data [23, 24]. maternal exposure to multiple toxins and the development of neonatal leukaemia has been extensively explored and an increased risk related to maternal marijuana use and alcohol 3. Transient Myeloproliferative Disorder consumption has been demonstrated [16, 17]. Therefore, a maternal diet high in flavonoids is suspected to increase the Transient myeloproliferative disorder (TMD) is clonal pro- risk of MLL-rearranged infant leukaemia [18]. liferation of megakaryoblasts, typically occurring in new- Congenital leukemia has a higher mortality rate than borns with Down syndrome (DS), with a range between any other form of congenital cancer; on the other hand, 4% and 6%. The first case was reported by Schunk and cases of spontaneous remission of AML have been described Lehman in 1954 [25]. Frequently, TMD disappears during in patients with AML [19]. These considerations raise the the first 3 months of life, but after a period of normal dilemma of deciding when to treat or to wait in case of a marrow morphology and peripheral blood count recovery, neonate presenting with congenital leukemia. a significant percentage of patients develop acute leukemia (most commonly AML) within the first 4 years of life. Because TMD appears to be restricted to patients with DS or trisomy 2.1. Clinical Presentation. The clinical presentation of neona- 21(T21)mosaicism,theoriginofAMLmustberelatedtoa tal leukemia can be variable. Patients are likely to present cytogenetic abnormality in chromosome 21 [26, 27]. TMD with hyperleukocytosis (65% of ALL patients and 49% is caused by cooperation between T21 and acquired somatic of AML patients), hepatosplenomegaly (80%), more often N-terminal truncating mutations in the key haematopoietic than enlarged spleen, central nervous system involvement transcription factor GATA1. These mutations, which are not (50%), and lymphadenopathy (24%). A severe complication leukaemogenic in the absence of T21, are found in almost of hyperleucocytosis is the leukostasis syndrome, in which one-third of neonates with DS. Analysis of primary human whitecellplugsareformedinthemicrovasculature,leading fetal liver haematopoietic cells and of human embryonic stem to cardiac failure and respiratory and neurological problems. cells demonstrates that T21 itself substantially alters human Neurological symptoms may include somnolence and coma, fetal haematopoietic development [28]. BioMed Research International 3

Table 2: Diagnostic tests in malignancies involving the skin, with differential diagnosis∗ ( in addition to the different causes of blueberry muffin baby listed in Table 1; WBC: white blood cell; Hb: Haemoglobin; PLTs: platelets).

∗ Neoplasia Diagnostic tests Differential diagnosis (i) Blood cell count (WBC > 50000; low Hb and PLTs) (i) Listeriosis (ii) Liver function tests (ii) Sepsis (iii) Coagulation studies (iii) Intrauterine Parvovirus infection (iv) Blood film (iv) Congenital HIV Leukemia (v)Bonemarrowaspiratewithmorphology, (v) Diamond Blackfan anemia immunophenotype (FAB L1, L2, M4, M5, and M7), and (vi) Extramedullary hematopoiesis karyotypes [t(4;11)(q21;q23); t(11;19)(q23;p13); (vii) Intrauterine or birth-related hypoxia t(9;11)(p21;q23); t(1;22)(p13;q13)] of blasts (viii) Neonatal lupus erythematosus (vi) Skin biopsy (i) Blood cell count (WBC 100000–50000 or normal; normal ordecreasedHb;anddecreasedPLTs) (ii) Liver function tests (increased bilirubin and/or transaminases) (i) Nonspecific changes associated with intrauterine (iii) Coagulation studies (abnormal coagulation) growth restriction and trisomies: neutropenia, (iv) Renal function tests thrombocytopenia, erythroblastosis, and TMD (v) Abdominal ultrasound polycythaemia (vi) Echocardiography (ii) Subtle myelodysplastic features: abnormal myeloid (vii) Chest X-ray cell granulation, giant platelets (viii) Blood film (ix) Bone marrow aspirate with morphology, immunophenotype (FAB M7), and karyotypes (GATA 1) of blasts (i) Hemophagocytic lymphohistiocytosis (ii) Familial erythrophagocytic lymphohistiocytosis (i) Blood cell count (normal or decreased Hb, RBC, and/or (iii) Infection-associated hemophagocytic syndrome PLTs) (iv) Hemangioendotheliomas (ii) Liver function tests (v) Extramedullary hematopoiesis LCH (iii) Coagulation studies (vi) Lymphatic malformations (iv) Chest X-ray (vii) Infantile myofibromatosis (v) Abdominal ultrasound (viii) Histiocytomas (vi) Skin biopsy (ix) Fibrosarcoma (x) Peripheral Primitive Neuroectodermal Tumor (i) Blood cell count (possible anemia and cytopenias) (ii) Lactate dehydrogenase levels (iii) Ferritin (iv) Liver and renal function tests (v) Urine catecholamines (increased homovanillic and (i) Benign cutaneous epithelioid Schwannoma NB vanillylmandelic acid) (ii) Ganglioneuroma (vi) Abdominal and cerebral ultrasound (iii) Ganglioneuroblastoma (vii) Chest X-ray (viii) If necessary, brain, neck, and chest MRI (ix) Skin biopsy with karyotypes (MYCN amplification) (x) Bone marrow aspirate (i) Urinalysis and renal function tests (ii) Abdominal ultrasound (i) Wilms’ tumor RT (iii) Chest X-ray (ii) Medulloblastoma (iv) Brain ultrasound (iii) Sarcomas (v) If necessary, brain MRI (iv) Hemangioma (vi) Skin biopsy with karyotypes (SWI/SNF at 22q11.2) RMS (i) Blood cell count (normal or decreased PLTs, anemia) (i) Wilms’ tumor (ii) Liver function tests (ii) Sarcomas

3.1. Clinical Presentation. TMD has variable presentation without leucocytosis. Babies with clinical symptoms most inthefetusandthenewbornfrommilddiseasetodis- commonly present with hepatomegaly, splenomegaly, jaun- seminated leukaemic infiltration and fulminant hepatic dice, and/or pleural/pericardial effusions. Less common fibrosis. Approximately 10–25% of newborns are asymp- presentations include liver fibrosis, ascites, and renal fail- tomatic and present with circulating blast cells, with or ure. 4 BioMed Research International

(a)

Figure 1: Skin involvement in the leukemia cutis, in a newborn on the first day of life. Randomly distributed subcutaneous nodules (arrows) on the chest (particularly in (a)) and on the trunk (c) with bluish infiltrates on the whole body surface (c) and particularly on the chest and the arm in (a) and (b). The blood cell rose until 167300/mcl, with prevalence of monocytes, low platelets count, and low hemoglobin levels. Infective pathologies have been excluded. The peripheral smear permitted the diagnosis of AML with positivity for CD45, CD33, CD14, CD13 heterogeneous, CD11b, and CD16 dull (AML-4). Genetic analysis revealed t(8;16)(p11;p13). After starting chemotherapy with aracytin, idarubicin, cytarabine, and etoposide, he died at 2 months of life.

The eruption, due to the cutaneous dissemination of the antiblastic agents and topical corticosteroids have been leukemic cells, is apparently more frequent in males and reported to accelerate the remission of the cutaneous picture. usually develops during the first days of life. Usually it is not associated with fever or respiratory distress, but infants present with high white blood cells count in the peripheral 4. Langerhans Cell Histiocytosis smear. Therefore, in newborns with diagnosed or suspected DS, presenting with cutaneous lesion, TMD needs to be con- The histiocytoses are a diverse group of diseases characterized sidered as a possible differential diagnosis. The recognition of by the infiltration and accumulation of histiocytes primarily this rare clinical entity can be important to differentiate other within the blood and tissues. Langerhans cell histiocytosis dermatoses requiring totally different management, such as (LCH) is the main histiocytosis occurring in the neonatal bacterial, viral, or fungal infections, histiocytosis, incontinen- period, followed in order by hemophagocytic lymphohisti- tia pigmenti, and, of course, leukemia cutis; therefore, further ocytosis (Figure 4) and juvenile xanthogranuloma [29–32]. specification of the clinical and histopathological findings is Neonatal LCH is rare, with a reported incidence of 1 per needed. 1 million newborns. Approximately 50% of cases involving Because of the possible evolution in acute leukemia DS multipleorgansareathigherriskofdisease-relatedmortality patientswithTMDhavetobestrictlymonitoredtoeventually [29, 33]. It is a rare disease marked by proliferation of start early therapy if needed. Langerhans-type cells, which shares immunophenotypic and ultrastructural similarities with antigen-presenting Langer- 3.2. Skin Involvement. This unusual form of vesiculopustu- hans cells of mucosal sites and skin [34]. lar eruption is due to the cutaneous dissemination of the The condition was initially described by Lichtenstein in leukemic cells and usually develops during the first days of 1953 and referred to as histiocytosis X to reflect the uncertain life. The primary lesions are papules, vesicles, or pustules that cause and often contrasting features of 3 clinical variants: canruptureandoriginateerosionandserosanguinouscrusts Letterer-Siwe disease, Hand-Schuller-Christian¨ disease, and (Figure 2); the face is almost invariably involved, but other eosinophilic granuloma [35]. Recently, the unifying term lesionsoftendevelopelsewhereandcanbeobservedover Langerhans cell histiocytosis has replaced histiocytosis X, any skin site, including scalp, palms, and soles. Cutaneous including congenital self-healing reticulohistiocytosis [36], histopathology shows intraepidermal vesicles or pustules Langerhans cell granulomatosis, and nonlipid reticuloen- (Figure 3) and a dermal mixed inflammatory infiltrate includ- dotheliosis. It is now the preferred terminology of any ing atypical mononuclear cells. Clearing without scarring disorder caused by proliferation of Langerhans cells. is spontaneous over few weeks or months, depending on Different oncogenic mutations have been involved inthe the course of the underlying haematological disorder, but LCH pathogenesis. BioMed Research International 5

(a) (b)

Figure 2: Lesions in a male neonate with Down syndrome, affected by TMD. Gradually, we observed evolution of the skin lesion, such as vesiculopustular eruption on the face (a) and vesiculopustular eruption on arm (b). A biopsy of the lesions showed mild acanthosis epidermis that was focally eroded. Within the superficial dermis and extending into the epidermis were large mononuclear cells with hyperchromatic nuclei and abundant eosinophilic cytoplasm consistent with immature myeloid cells. Immunohistochemical stains showed strong immunoreactivity of these cells with myeloperoxidase and a megakaryocytic cell marker LAT (linker for activation of T cells), confirming that the cells were of myeloid and megakaryoblastic origin (Figure 3). The histologic findings supported the clinical suspicion of a TMD.

(a) (b)

Figure 3: Image of cutaneous lesion biopsy, low-power magnification showing intraepidermal pustule and dermal perivascular infiltrates (hematoxylin-eosin staining) 10x (a). Immunoreaction for LAT (linker for activation of T cells): the cells in the dermis look like atypical blasts of myeloid and megakaryocytic lineage (b).

BRAFV600E has been detected in more than half of the nodules, other typical cutaneous lesions are scaly, erythe- cases of LCH in adult age. BRAFV600E mutation results matous, seborrhoea-like eruptions of brown to red papules. in constitutive activation of the mitogen-activated protein Superficial ulcerations within these lesions are also described kinase (MAPK) pathway; it has been observed that the MAPK resulting in weeping lesions suggestive of eczema. pathway is also activated in cases of LCH without mutations In addition, patients with multisystem disease have of BRAF [37]. Mutations in MAP2K1, which encodes the involvement of at least one other system, including the dual-specificity kinase MEK1 protein in the MAPK pathway, bone, lymph nodes, the central nervous system, the eye, have been identified in 27.5% of cases with LCH, thus the gastrointestinal tract, the bone marrow, or solid organs. explaining MAPK pathway activation in the absence of the Patients may present with pain, dyspnoea, and failure to BRAF mutation. MAP2K1 and BRAF mutations were found thrive [33, 41]. to be mutually exclusive, as would be expected since MEK1 is directly downstream BRAF within the MAPK pathway [38]. 4.2. Skin Involvement. Various skin findings are character- Additional cases of other mutations in the MAPK genes have isticofLCH.Themostcommonlesionsareredbrown also been reported, including ARAF and ERBB3 [39, 40]. papules or nodules that may be pseudo vesicular and often crusty. They can be associated with secondary erosion or 4.1. Clinical Presentation. LCH is characterized by the infil- hemorrhage. In addition, erythematous, scaly dermatitis tration and accumulation of histiocytes and other immune involving the scalp, posterior auricular regions, perineum, effector cells within various tissues. In addition to blue and/oraxillaecanbeseen[42,43]. 6 BioMed Research International

disease), were the first germline mutations to be identified in NB predisposition [50–52]. More recently, whole exome and whole genome sequencing analyses have identified loss-of- function mutations/deletions in chromatin modifiers including ATRX, ARID1A,andARID1B.However,ATRX mutations were more common in patients over 5 years of age and no mutations/deletions were identified in the youngest age group (<18 months) [53]. In neonates and infants aged <18 months, NB has a good prognosis as most have favourable biological characteristics and may undergo spontaneous regression even if metastatic. Figure 4: A reticulated, maculopapular rash in the lower limb Diagnosis can occasionally be performed during pregnancy in a term male affected by haemophagocytic lymphohistiocytosis. by fetal ultrasound [54]. He presented with fever associated with generalized rash which disappearedinthefirstweekoflifeandappearedagainon30thday. 5.1. Clinical Presentation. Multiple nodules occur in more Blood test analysis showed values indicative of liver failure; blood than one-third of patients and are characteristic of stage cell count revealed low platelets values. The abdominal ultrasound scan showed thinly irregular liver parenchyma. Hypoplasia of gran- 4S disease. Stage 4S NB is defined by the International ulocyte and erythroblastic and megakaryocytic lineages (<5% each) Neuroblastoma Staging System (INSS) as metastatic disease with haematopoietic cell phagocytosis by mature macrophages were with metastases confined to skin, liver, and/or bone marrow present in the bone marrow aspirate. High-dose glucocorticoids, in infants younger than one. According to the Stage Interna- cyclosporine, and etoposide were started. Despite the therapy, there tional Neuroblastoma Risk Group, based on clinical criteria has been gradual deterioration of the clinical condition until coma, andimage-definedriskfactors,theageforstage4S(also respiratory failure, intractable hypotension, and coagulopathy led to called stage MS) has been extended to 18 months [55, 56]. death despite intensive vital support. Stage 4S (or MS) can also present with rapid enlarging, diffusively involved liver (“Pepper Syndrome”), which may cause respiratory compromise, renal impairment, and bowel Askinbiopsyprovidesarapidandaccessiblemeans dysfunction together with coagulopathy. NB can also cause tosecurethediagnosis.ApresumptivediagnosisofLCH spinal cord compression in the neonate, with bladder and may be made based upon light microscopic findings and a bowel dysfunction as well as motor impairment [57, 58]. compatible clinical picture, but a definitive diagnosis requires that lesional cells exhibit positive staining with S-100 and 5.2. Skin Involvement. The typical cutaneous eruption is CD1a, and the identification of Birbeck granules upon elec- represented by bluish cutaneous nodules with a characteristic tron microscopy is necessary [44]. blanch response to palpation that leaves surrounding rim of erythema. Ocular signs such as “raccoon eyes,” periorbital ecchymosis, or heterochromia iridis may be present [59]. 5. Neuroblastoma Histological examination of the cutaneous nodules Neuroblastoma (NB) is the most common neonatal malig- reveals a uniform, small cell, malignant tumor with or with- nancy. The Automated Childhood Cancer Information Sys- out Homer-Wright pseudo rosette formation. Molecularly, tem (ACCIS) revealed an incidence of 27% of cancers cases NB stage 4S is characterised by near triploidy and absence of in Europe between 1978 and 1997, with the highest incidence genetic alterations. Structural genetic changes characteristic inthefirstyearoflife[45].Onestudyreportedthat16% of stage 4 NB (e.g., segmental chromosomal alterations, of infant neuroblastomas were diagnosed during the first MYCN amplification, and ALK mutations) are not present in month of life and 42% during the first 3 months [46]. children with NB stage 4S, but if they are present, the tumor Stage 4S represents approximately 7–10% of all cases of NB. behaves as if it was a stage 4 NB. Most neuroblastomas occur sporadically; familial NB per se is rare and concerns only approximately 1% of all cases 6. Rhabdoid Tumor [47]. Different genomic mutations have been identified in thelastdecades.ThemostimportantalterationisMYCN Rhabdoid tumor (RT) was distinguished from Wilms’ tumor amplification, which is strongly correlated with advanced in the 1970s. This highly malignant neoplasm is characterized disease, drug resistance, and poor outcome [48]. The ampli- by early metastases and a high mortality rate. The tumor fication of MYCN and the subsequent overexpression of occurs in the perinatal period during the first year of life the protein directly contribute to tumorigenesis. MYCN and occasionally in older children. According to the origin oncogene is predominantly expressed in the developing tissue the RT family is divided into 3 categories, with an peripheral neural crest inducing proliferation and migration, approximately equal distribution: (1) primary central nervous with decreased levels associated with terminal differentiation system lesions, defined as atypical teratoid/rhabdoid tumor, [49]. Missense mutations of Paired Homeobox 2b (PHOX2B) associated with a low incidence of metastases outside the on chromosome 4p, frequently associated with other neural central nervous system, (2) primary renal lesions, and (3) crest disorders or malignancies (Ondine’s and Hirschsprung’s primary soft tissue lesions. The latter typically relates to BioMed Research International 7

(a) (b)

Figure 5: Nodular lesions diffused on the body in a term baby with alveolar rhabdomyosarcoma. Particular image of the chest and right arm (a) and the legs (b). An incisional biopsy of the cutis showed aggregates of atypical small spindle cells interrupted by fibrovascular septae and collagen bands resembling the lung alveoli. Bone marrow biopsy and total body CT scan demonstrated the presence of pulmonary, pleural, and pericardial metastasis. Chemotherapy was started, without a clinical improvement. The baby died at 26 days of life because of renal insufficiency and heart failure. widespread metastatic disease [60–64]. The incidence of seen by electron microscopy [64, 73]. Moreover, the presence metastatic RT/atypical RT is not defined; however, European of a mutation of the hsNF5/INI1 gene located in chromosome data indicate 0.1 to 0.5 per million children per year. Rhab- 22q11 is helpful in establishing the diagnosis. doid tumors not involving the kidney are very rare [65]. When RT occurs in utero, it is more likely to present at birth 7. Rhabdomyosarcoma with multiple metastases and a rapidly progressive, downhill clinical course ending in early death. Metastatic disease is Rhabdomyosarcoma (RMS), the most common soft tissue present in more than half of the neonates at the time of sarcoma, represents 4–8% of all malignant solid tumors diagnosis. in childhood, with few cases described in infants younger The vast majority of RTs demonstrate abnormalities in than one month of life [74–76]. RMS may be congenital. chromosome 22. These abnormalities are characterized by the RMS is traditionally subdivided into embryonal, alveolar, and loss of function of a member of the SWI/SNF chromatin- pleomorphic. The overall prognosis of perinatal RMS is poor, remodeling complex located at 22q11.2, known with different with expected survival around 40% [77]. names (hSNF5, INI1, BAF47,orSMARCB1)[66].InRT, genetic mutations are characterized by somatically acquired 7.1. Clinical Presentation. Most cases of RMS occur in the biallelic inactivating truncating mutations within the tumor head and neck region (29.3%) and in the genitourinary tract cells, associated or not with a predisposing germline mutation (26.3%) [78]. Multiple skin nodules can be observed. In [67, 68]. addition, neonates can be manifested with haemangioma-like vascular lesions and an abdominal mass; hepatosplenomegaly 6.1. Clinical Presentation. RT may present in the skin, partic- and thrombocytopenia can be associated. Neonatal Alveolar ularly in the head and neck area, as a solitary primary tumor RMS (Figure 5) frequently manifests itself with multiple skin or as metastatic skin nodules. RTs have to be suspected in case lesions associated with brain metastasis. of haematuria with or without hypercalcemia [69]. Neonates can also manifest an increase in cranial circumference due 7.2. Skin Involvement. RMS is clinically manifested as multi- to hydrocephalus, seizures, and irritability because of CNS ple papulonodular lesions and subcutaneous nodules or as a involvement. Moreover, vomiting, anaemia, fever, and respi- single papule or nodule, which grows slowly. The lesions may ratory distress can be present [70]. present as soft cherry-red color nodules or as firm violaceous subcutaneous nodules [79]. 6.2. Skin Involvement. Few case reports describe the cuta- Most cases of RMS skin metastasis are of the alveolar neous involvement due to RT [70–72]. Typically primary subtype, with small- to medium-sized darkly stained cells central nervous system lesions can manifest themselves as with rounded nuclei and scant cytoplasm arranged in alveolar one or several lightly erythematous papulonodular lesions. pattern. It is often very difficult to differentiate the skin They seem to be verrucous, as observed with verrucous metastases of RMS from other skin neoplasms that are hamartomas. composed of spindle or round cells, such as lymphoma, NB, The histologic diagnosis of RT is based on the identifica- andEwing’ssarcoma[80].Theuseofimmunohistochemistry tion of the characteristic RT cell, which consists of a round may provide additional information to aid in differential vesicular nucleus, a prominent nucleolus, and round to oval diagnosis. Cytogenetic analysis plays an important role in eosinophilic inclusion containing intermediate filaments as confirming the diagnosis, but only for alveolar RMS because 8 BioMed Research International there are no specific cytogenetic or molecular markers for 1743–1795, Mosby Elsevier, Philadelphia, Pa, USA, 2nd edition, embryonal RMS. Alveolar RMS is associated with specific 2007. translocation, t(2;13)(q37;q14) or its variant t(1;13)(p36;q14) [7] S. Weitzman and R. Grant, “Neonatal oncology: diagnostic and [81]. therapeutic dilemmas,” Seminars in Perinatology,vol.21,no.1, pp. 102–111, 1997. 8. Other Tumors [8] T. S. Wright, “Cutaneous manifestations of malignancy,” Cur- rent Opinion in Pediatrics, vol. 23, no. 4, pp. 407–411, 2011. Few cases of other neoplasms with cutaneous involvement [9] M. H. Van der Linden, S. Creemers, and R. Pieters, “Diagnosis as first presentation in the first days of life are described. and management of neonatal leukaemia,” Seminars in Fetal and The rare Peripheral Primitive Neuroectodermal Tumor is Neonatal Medicine,vol.17,no.4,pp.192–195,2012. characterized by the presence of small blue cells in the CNS [10] H. Isaacs Jr., “Fetal and neonatal leukemia,” Journal of Pediatric andinperipheraltissues[80].Neonatalchoriocarcinoma, Hematology/Oncology,vol.25,no.5,pp.348–361,2003. an aggressive malignant tumor of trophoblastic cells, with [11] M. P. Bas Suarez,´ J. Lopez´ Brito, C. Santana Reyes, M. Gresa cutaneous metastasis has been described in four neonates Munoz,˜ R. Diaz Pulido, and J. C. Lodos Rojas, “Congenital acute [82]. The infantile myofibromatosis, the most common lymphoblastic leukemia: a two-case report and a review of the fibrous tumor of infancy, is characterized by the presence of literature,” European Journal of Pediatrics,vol.170,no.4,pp.531– myofibroblastic cells that involve the skin, soft tissue, bones, 534, 2011. and internal organs [83]. [12] D. Bresters, A. C. Reus, A. J. Veerman, E. R. van Wering, A. van der Does-van den Berg, and G. J. Kaspers, “Congenital leukaemia: the Dutch experience and review of the literature,” 9. Conclusion British Journal of Haematology,vol.117,no.3,pp.513–524,2002. [13] H. Chang, J. Brandwein, Q.-L. Yi, K. Chun, B. Patterson, and Theblueberrymuffinrashaswellasnodules,bruising, B. Brien, “Extramedullary infiltrates of AML are associated and crusts may all indicate serious underlying diseases and with CD56 expression, 11q23 abnormalities and inferior clinical sometimes may represent the first clinical manifestation of outcome,” Leukemia Research, vol. 28, no. 10, pp. 1007–1011, rare neoplasms in the newborn infant. Clinical and laboratory 2004. findings can be limited and clinic-pathological correlation is [14] M. Y. Hurley, G. K. Ghahramani, S. Frisch et al., “Cutaneous critical. A definitive diagnosis can be obtained only by means myeloid sarcoma: natural history and biology of an uncom- of a skin biopsy. Treatment options pose challenges because mon manifestation of acute myeloid leukemia,” Acta Dermato- of the particular vulnerability of the affected neonates. Venereologica,vol.93,no.3,pp.319–324,2013. Neonatal cancer raises diagnostic, therapeutic, and ethical [15] E. A. Coenen, C. M. Zwaan, D. Reinhardt et al., “Pediatric issues; management requires a multidisciplinary approach. acute myeloid leukemia with t(8;16)(p11;p13), a distinct clinical The prognosis depends on the type of cancer identified, the and biological entity: a collaborative study by the International- optimal timing of the therapy, and the subsequent follow-up. Berlin-Frankfurt-Munster AML-study group,” Blood,vol.122, no.15,pp.2704–2713,2013. Conflict of Interests [16]X.-O.Shu,J.A.Ross,T.W.Pendergrass,G.H.Reaman,B. Lampkin, and L. L. Robison, “Parental alcohol consumption, The authors declare that there is no conflict of interests cigarette smoking, and risk of infant leukemia: a Childrens regarding the publication of this paper. Cancer Group study,” Journal of the National Cancer Institute, vol.88,no.1,pp.24–31,1996. [17] L. L. Robison, J. D. Buckley, A. E. Daigle et al., “Maternal References drug use and risk of childhood nonlymphoblastic leukemia among offspring. An epidemiologic investigation implicating [1] S. E. Parkes, K. R. Muir, L. Southern, A. H. Cameron, P. J. marijuana (a report from the childrens cancer study group),” Darbyshire, and M. C. G. Stevens, “Neonatal tumours: a thirty- Cancer,vol.63,no.10,pp.1904–1911,1989. year population-based study,” Medical and Pediatric Oncology, vol. 22, no. 5, pp. 309–317, 1994. [18] J. A. Ross, “Dietary flavonoids and the MLL gene: a pathway to infant leukemia?” Proceedings of the National Academy of [2] J. M. Birch and V. Blair, “The epidemiology of infant cancers,” Sciences of the United States of America,vol.97,no.9,pp.4411– British Journal of Cancer,vol.66,pp.S52–S54,1992. 4413, 2000. [3]S.W.Moore,D.Satge,´ A. J. Sasco, A. Zimmermann, and J. [19]H.vandenBerg,A.H.N.Hopman,K.C.Kraakman,and Plaschkes, “The epidemiology of neonatal tumours: report of an D. De Jong, “Spontaneous remission in congenital leukemia international working group,” Pediatric Surgery International, is not related to (MOSAIC) trisomy 21: case presentation and vol.19,no.7,pp.509–519,2003. literature review,” Pediatric Hematology and Oncology,vol.21, [4] H. Isaacs Jr., “Cutaneous metastases in neonates: a review,” no. 2, pp. 135–144, 2004. Pediatric Dermatology,vol.28,no.2,pp.85–93,2011. [20] P. Porcu, S. Farag, G. Marcucci, S. R. Cataland, M. S. Kennedy, [5] H. R. Isaacs Jr., “Tumors and tumorlike conditions of the skin,” and M. Bissell, “Leukocytoreduction for acute leukemia,” Ther- in Tumors of the Fetus and Infant: An Atlas,pp.113–135,Springer, apeutic Apheresis,vol.6,no.1,pp.15–23,2002. New York, NY, USA, 2002. [21] M. O. Arcasoy and P.G. Gallagher, “Hematologic disorders and [6] H. R. Isaacs Jr., “Pathology of skin diseases,”in Potter’s Pathology nonimmune hydrops fetalis,” Seminars in Perinatology,vol.19, of the Fetus, Infant and Child,E.Gilbert-Barness,Ed.,vol.2,pp. no. 6, pp. 502–515, 1995. BioMed Research International 9

[22] T. A. Longacre and B. R. Smoller, “Leukemia cutis. Analysis [38] N. A. Brown, L. V.Furtado, B. L. Betz et al., “High prevalence of of 50 biopsy-proven cases with an emphasis on occurrence somatic MAP2K1 mutations in BRAF V600E-negative Langer- in myelodysplastic syndromes,” American Journal of Clinical hans cell histiocytosis,” Blood,vol.124,no.10,pp.1655–1658, Pathology,vol.100,no.3,pp.276–284,1993. 2014. [23]E.Sotiriou,A.Manousari,Z.Apalla,I.Papagarifallou,andD. [39] R. Chakraborty, O. A. Hampton, X. Shen et al., “Mutually Ioannides, “Aleukaemic congenital leukaemia cutis: a critical exclusive recurrent somatic mutations in MAP2K1 and BRAF primary sign of systemic disease,” Acta Dermato-Venereologica, support a central role for ERK activation in LCH pathogenesis,” vol.91,no.2,pp.203–204,2011. Blood,vol.124,no.19,pp.3007–3015,2014. [24]I.H.Zhang,L.T.Zane,B.S.Braun,J.MaizeJr.,S.Zoger, [40] D. S. Nelson, W. Quispel, G. Badalian-Very et al., “Somatic and M. L. Loh, “Congenital leukemia cutis with subsequent activating ARAF mutations in Langerhans cell histiocytosis,” development of leukemia,” Journal of the American Academy of Blood,vol.123,no.20,pp.3152–3155,2014. Dermatology,vol.54,no.2,pp.S22–S27,2006. [41] S. L. Stein, A. S. Paller, P. R. Haut, and A. J. Mancini, “Langer- [25] G. J. Schunk and W. L. Lehman, “Mongolism and congenital hans cell histiocytosis presenting in the neonatal period: a leukemia,” The Journal of the American Medical Association,vol. retrospective case series,” Archives of Pediatrics and Adolescent 155, no. 3, pp. 250–251, 1954. Medicine,vol.155,no.7,pp.778–783,2001. [42]A.Paller,A.J.Mancini,A.Hurwitz,S.Paller,andA.Mancini, [26] G. V. Massey, A. Zipursky, M. N. Chang et al., “A prospective “Histiocytoses and malignant skin diseases,”in Hurwitz Clinical study of the natural history of transient leukemia (TL) in Pediatric Dermatology: A Textbook of Skin Disorders of Child- neonates with Down syndrome (DS): Children’s Oncology hood and Adolescence, pp. 245–263, WB Saunders, Philadelphia, Group (COG) study POG-9481,” Blood,vol.107,no.12,pp. Pa,USA,3rdedition,2011. 4606–4613, 2006. [43]M.J.Ehrhardt,S.R.Humphrey,M.E.Kelly,Y.E.Chiu,andS. [27] G. V. Massey, “Transient leukemia in newborns with Down S. Galbraith, “The natural history of skin-limited Langerhans syndrome,” Pediatric Blood and Cancer,vol.44,no.1,pp.29– cell histiocytosis: a single-institution experience,” Journal of 32, 2005. Pediatric Hematology/Oncology,vol.36,no.8,pp.613–616,2014. [28] I. Roberts and S. Izraeli, “Haematopoietic development and [44] E. K. Satter and W. A. High, “Langerhans cell histiocytosis: leukaemia in Down syndrome,” British Journal of Haematology, a review of the current recommendations of the Histiocyte vol. 167, no. 5, pp. 587–599, 2014. Society,” Pediatric Dermatology, vol. 25, no. 3, pp. 291–295, 2008. [29] H. Isaacs Jr., “Fetal and neonatal histiocytoses,” Pediatric Blood [45]C.Spix,G.Pastore,R.Sankila,C.A.Stiller,andE.Steliarova- and Cancer,vol.47,no.2,pp.123–129,2006. Foucher, “Neuroblastoma incidence and survival in European [30]D.S.Morrel,M.A.Pepping,J.P.Scott,N.B.Esterly,andB.A. children (1978–1997): report from the Automated Childhood Drolet, “Cutaneous manifestation of hemophagocitic lympho- Cancer Information System project,” European Journal of Can- histiocytosis,” Archives of Dermatology,vol.138,pp.1208–1212, cer, vol. 42, no. 13, pp. 2081–2091, 2006. 2002. [46] S. Dhir and K. Wheeler, “Neonatal neuroblastoma,” Early [31] O. Danhaive, M. Caniglia, R. Devito, F. Piersigilli, C. Corchia, Human Development, vol. 86, no. 10, pp. 601–605, 2010. and C. Auriti, “Neonatal liver failure and haemophagocytic [47]G.Schleiermacher,H.Rubie,O.Hartmannetal.,“Treatmentof lymphohistiocytosis caused by a new perforin mutation,” Acta stage 4S neuroblastoma—report of 10 years’ experience of the Paediatrica,vol.99,no.5,pp.778–780,2010. French Society of Paediatric Oncology (SFOP),” British Journal of Cancer,vol.89,no.3,pp.470–476,2003. [32]D.R.Freyer,R.Kennedy,B.C.Bostrom,G.Kohut,andL.P. Dehner, “Juvenile xanthogranuloma: forms of systemic disease [48] G. M. Brodeur, “Neuroblastoma: biological insights into a and their clinical implications,” Journal of Pediatrics,vol.129,no. clinical enigma,” Nature Reviews Cancer,vol.3,no.3,pp.203– 2, pp. 227–237, 1996. 216, 2003. [49]W.A.Weiss,K.Aldape,G.Mohapatra,B.G.Feuerstein,andJ. [33] M. Minkov, H. Prosch, M. Steiner et al., “Langerhans cell M. Bishop, “Targeted expression of MYCN causes neuroblas- histiocytosis in neonates,” Pediatric Blood and Cancer,vol.45, toma in transgenic mice,” The EMBO Journal,vol.16,no.11,pp. no. 6, pp. 802–807, 2005. 2985–2995, 1997. [34] R. Jaffe, L. M. Weiss, and F. Facchetti, “Tumours derived [50] N. Clausen, P. Andersson, and N. Tommerup, “Familial occur- from Langerhans cells,” in WHO Classification of Tumours rence of neuroblastoma, von Recklinghausen’s neurofibromato- of Haematopoietic and Lymphoid Tissues,S.H.Swerdlow,E. sis, Hirschsprung’s agangliosis and jaw-winking syndrome,” Campo, N. L. Harris et al., Eds., vol. 2 of World Health Acta Paediatrica Scandinavica,vol.78,no.5,pp.736–741,1989. Organization Classification of Tumours, pp. 358–360, IARC [51] T. Rohrer, D. Trachsel, G. Engelcke, and J. Hammer, “Con- Press, Lyon, France, 4th edition, 2008. genital central hypoventilation syndrome associated with [35]A.C.Chu,R.H.Champion,J.L.Burton,andD.A.Burns, Hirschsprung’s disease and neuroblastoma: case of multiple “Histiocytoses,” in Rook’s Textbook of Dermatology,S.M. neurocristopathies,” Pediatric Pulmonology,vol.33,no.1,pp.71– Breathnach, Ed., pp. 2311–2336, Blackwell Science, London, UK, 76, 2002. 1998. [52] D. Trochet, L. M. O’Brien, D. Gozal et al., “PHOX2B geno- [36] M. Inuzuka, K. Tomita, Y. Tokura, and M. Takigawa, “Congen- type allows for prediction of tumor risk in congenital central ital self-healing reticulohistiocytosis presenting with hemor- hypoventilation syndrome,” The American Journal of Human rhagic bullae,” Journal of the American Academy of Dermatology, Genetics, vol. 76, no. 3, pp. 421–426, 2005. vol.48,no.5,supplement,pp.S75–S77,2003. [53] N.-K. V. Cheung, J. Zhang, C. Lu et al., “Association of age [37] G. Badalian-Very, J.-A. Vergilio, B. A. Degar et al., “Recurrent at diagnosis and genetic mutations in patients with neuroblas- BRAF mutations in Langerhans cell histiocytosis,” Blood,vol. toma,” TheJournaloftheAmericanMedicalAssociation,vol.307, 116, no. 11, pp. 1919–1923, 2010. no.10,pp.1062–1071,2012. 10 BioMed Research International

[54] S. Acharya, S. Jayabose, S. J. Kogan et al., “Prenatally diagnosed [71] C. Hsueh and T.-T. Kuo, “Congenital malignant rhabdoid tumor neuroblastoma,” Cancer, vol. 80, no. 2, pp. 304–310, 1997. presenting as a cutaneous nodule: report of 2 cases with review [55] G. M. Brodeur, J. Pritchard, F. Berthold et al., “Revisions of the of the literature,” Archives of Pathology and Laboratory Medicine, international criteria for neuroblastoma diagnosis, staging, and vol. 122, no. 12, pp. 1099–1102, 1998. response to treatment,” Journal of Clinical Oncology, vol. 11, no. [72] A. Dominey, A. S. Paller, and F. Gonzalez-Crussi, “Congenital 8, pp. 1466–1477, 1993. rhabdoid sarcoma with cutaneous metastases,” Journal of the [56]T.Monclair,G.M.Brodeur,P.F.Ambrosetal.,“TheInterna- American Academy of Dermatology, vol. 22, no. 5, pp. 969–974, tional Neuroblastoma Risk Group (INRG) staging system: an 1990. INRG Task Force report,” Journal of Clinical Oncology,vol.27, [73]D.A.Weeks,J.B.Beckwith,G.W.Mierau,andD.W.Luckey, no. 2, pp. 298–303, 2009. “Rhabdoid tumor of kidney: a report of 111 cases from the [57] M. M. van Noesel, “Neuroblastoma stage 4S: a multifocal stem- National Wilms’ Tumor Study Pathology Center,” The American cell disease of the developing neural crest,” The Lancet Oncology, Journal of Surgical Pathology,vol.13,no.6,pp.439–458,1989. vol. 13, no. 3, pp. 229–230, 2012. [74] S. V. Godambe and J. Rawal, “Blueberry muffin rash asa [58] J. P. H. Fisher and D. A. Tweddle, “Neonatal neuroblastoma,” presentation of alveolar cell rhabdomyosarcoma in a neonate,” Seminars in Fetal & Neonatal Medicine,vol.17,no.4,pp.207– Acta Paediatrica,vol.89,no.1,pp.115–117,2000. 215, 2012. [75] F.Piersigilli, O. Danhaive, and C. Auriti, “Blueberry muffin baby [59] V. Mehta, C. Balachandran, and V. Lonikar, “Blueberry muffin due to alveolar rhabdomyosarcoma with cutaneous metastasis,” baby: a pictoral differential diagnosis,” Dermatology Online ArchivesofDiseaseinChildhood:FetalandNeonatalEdition,vol. Journal,vol.14,article8,2008. 95, article F461, 2010. [60] F. V. White, L. P. Dehner, D. A. Belchis et al., “Congenital [76]N.Kitagawa,J.Arata,Y.Ohtsuki,K.Hayashi,Y.Oomori,andT. disseminated malignant rhabdoid tumor: a distinct clinico- Tomoda, “Congenital alveolar rhabdomyosarcoma presenting pathologic entity demonstrating abnormalities of chromosome as a blueberry muffin baby,” JournalofDermatology,vol.16,no. 22q11,” The American Journal of Surgical Pathology,vol.23,no. 5, pp. 409–411, 1989. 3, pp. 249–256, 1999. [77] I. Sultan, M. Casanova, U. Al-Jumaily, C. Meazza, C. Rodriguez- [61] M. Sajedi, J. E. A. Wolff, R. M. Egeler et al., “Congenital Galindo, and A. Ferrari, “Soft tissue sarcomas in the first year of extrarenal non-central nervous system malignant rhabdoid life,” European Journal of Cancer,vol.46,no.13,pp.2449–2456, tumor,” Journal of Pediatric Hematology/Oncology,vol.24,no. 2010. 4, pp. 316–320, 2002. [78] C.-Y.Yang, I.-H. Shih, C.-M. Wang, and H.-L. Chan, “Embryon- [62] L. B. Rorke, R. J. Packer, and J. A. Biegel, “Central nervous alrhabdomyosarcoma with skin invasion mimicking a perianal system atypical teratoid/rhabdoid tumors of infancy and child- abcess,” Dermatologica Sinica,vol.18,pp.219–222,2000. hood: definition of an entity,” Journal of Neurosurgery,vol.85, [79] R. Grundy, J. Anderson, M. Gaze et al., “Congenital alveolar no. 1, pp. 56–65, 1996. rhabdomyosarcoma: clinical and molecular distinction from [63] C. Sotelo-Avila, F. Gonzalez-Crussi, D. deMello et al., “Renal alveolar rhabdomyosarcoma in older children,” Cancer,vol.91, and extrarenal rhabdoid tumors in children: a clinicopathologic no.3,pp.606–612,2001. study of 14 patients,” Seminars in Diagnostic Pathology,vol.3,no. [80] C. Paley, E. Valderrama, M. Garcia, and G. Karayalcin, “Con- 2, pp. 151–163, 1986. genital peripheral neuroectodermal tumor presenting as dis- [64]A.R.M.Cobb,N.J.Sebire,J.Anderson,andD.Dunaway, seminated cutaneous disease,” Journal of Pediatric Hematol- “Congenital malignant rhabdoid tumor of the scalp,” Journal of ogy/Oncology,vol.18,p.447,1996. Cranio-Maxillofacial Surgery,vol.40,no.8,pp.e258–e260,2012. [81] C. De Giovanni, L. Landuzzi, G. Nicoletti, P.-L. Lollini, and P. [65] H. Reinhard, J. Reinert, R. Beier et al., “Rhabdoid tumors Nanni, “Molecular and cellular biology of rhabdomyosarcoma,” in children: prognostic factors in 70 patients diagnosed in Future Oncology,vol.5,no.9,pp.1449–1475,2009. Germany,” Oncology Reports,vol.19,no.3,pp.819–823,2008. [82] E. J. Johnson, P. M. Crofton, J. O’Neill et al., “Infantile chori- [66] C. W.M. Roberts and J. A. Biegel, “The role of SMARCB1/INI1 in ocarcinoma treated with chemotherapy alone,” Medical and development of rhabdoid tumor,” Cancer Biology and Therapy, Pediatric Oncology,vol.41,no.6,pp.550–557,2003. vol. 8, no. 5, pp. 412–416, 2009. [83] C. Auriti, M. W. Kieran, G. Deb, R. Devito, L. Pasquini, and O. [67] J. A. Biegel, L. Tan, F. Zhang, L. Wainwright, P. Russo, and Danhaive, “Remission of infantile generalized myofibromatosis L. B. Rorke, “Alterations of the hSNF5/INI1 gene in central after interferon alpha therapy,” Journal of Pediatric Hematol- nervous system atypical teratoid/rhabdoid tumors and renal ogy/Oncology,vol.30,no.2,pp.179–181,2008. and extrarenal rhabdoid tumors,” Clinical Cancer Research,vol. 8, no. 11, pp. 3461–3467, 2002. [68] J. A. Biegel, B. Fogelgren, L. M. Wainwright, J.-Y. Zhou, H. Bevan, and L. B. Rorke, “Germline INI1 mutation in a patient with a central nervous system atypical teratoid tumor and renal rhabdoid tumor,” Genes, Chromosomes and Cancer,vol.28,no. 1,pp.31–37,2000. [69]A.M.Amar,G.Tomlinson,D.M.Green,N.E.Breslow,andP. A. de Alarcon, “Clinical presentation of rhabdoid tumors of the kidney,” Journal of Pediatric Hematology/Oncology,vol.23,no. 2, pp. 105–108, 2001. [70] H. Isaacs Jr., “Fetal and neonatal rhabdoid tumor,” Journal of Pediatric Surgery,vol.45,no.3,pp.619–626,2010. Hindawi Publishing Corporation BioMed Research International Volume 2015, Article ID 321896, 8 pages http://dx.doi.org/10.1155/2015/321896

Research Article Surgical Treatment and Prognosis of Angiosarcoma of the Scalp: A Retrospective Analysis of 14 Patients in a Single Institution

Jun Ho Choi, Kyung Chan Ahn, Hak Chang, Kyung Won Minn, Ung Sik Jin, and Byung Jun Kim Department of Plastic and Reconstructive Surgery, Seoul National UniversityHospital,SeoulNationalUniversityCollegeofMedicine, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea

Correspondence should be addressed to Byung Jun Kim; [email protected]

Received 11 September 2015; Revised 9 November 2015; Accepted 15 November 2015

Academic Editor: Elisabeth Roider

Copyright © 2015 Jun Ho Choi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Objective. We describe specific surgical methods for angiosarcoma regarding extent of resection and reconstructive options and assess their effect on patients’ prognosis. Patients and Methods. We retrospectively examined 14 patients undergoing treatment for angiosarcoma of the scalp at our institute between January 2000 and June 2015. Surgical treatment comprised wide excision of the tumor and reconstruction using a free flap with skin graft. Kaplan-Meier survival analysis was used to assess the survival parameters. Univariate and multivariate analyses were performed to evaluate the association between risk factors and outcome parameters. Results. Mean patient age at diagnosis was 69 years, and the mean follow-up period was 17 months. The overall 5- and 2-year survival rates were 15% and 75%, respectively, whereas the 5- and 2-year disease-free survival rates were 7.7%and 38.7%, respectively. The mean survival duration was 32 months. Metastatic tumor dissemination to the lung or brain was closely associated with the major cause of death. Only a deep excision margin was significantly related to the recurrence rate. Conclusions.Casesof angiosarcoma had a poor prognosis despite the aggressive treatments. Sufficient resection margins are essential for controlling local recurrence. The effect of multidisciplinary approaches needs to be explored.

1. Introduction radiotherapy is broadly accepted as the standard treatment [2, 6, 14]. As concrete surgical strategies have still not been estab- Angiosarcoma (AS) is rare vascular neoplasm affecting the lished, in the present report, we describe surgical methods endothelial cells of blood vessel. Cutaneous AS usually for AS in terms of the extent of resection and reconstructive develops in the face or scalp [1, 2]. This tumor predominantly options and assess their effect on patient prognosis. develops in elderly people and more frequently affects men than women [2–4]. AS accounts for less than 2% of all cases 2. Materials and Methods of soft-tissue sarcoma and less than 1% of cases of head and neck cancer [5–7]. The etiology of AS has not been fully 2.1. Study Design and Patients. The subjects of the present understood, but several reports have shown some positive study are patients with AS of the scalp, who were treated at association with chronic lymphedema [8] or prior irradiation our institute between January 2000 and June 2015, and their [9,10].Itisahighlyaggressivemalignanttumorwithahigh medical records were retrospectively reviewed. All the proce- rate of locoregional recurrence and tends to metastasize at dures in the present study were approved by our institutional an earlier stage. Although various active treatment options review board (IRB number: 1508-088-695). Patients in the have been adopted, the associated prognosis remains poor. study were primarily diagnosed with AS, and there was no Many studies have indicated that the 5-year survival rate of prior history of radiotherapy. Patients with AS located at any AS ranges between 10% and 50% [4, 7, 11–13]. No common other site except the scalp were excluded from the study. The consensus has been reached regarding the optimal treatment reason for such exclusion was that the scalp has a unique [6]. However, a multimodality approach of wide excision and anatomy which significantly affects the surgical decisions. 2 BioMed Research International

(a) (b)

Figure 1: Intraoperative photographs during wide excision of angiosarcoma of the scalp. A wide soft-tissue defect had been made after wide excision of the tumor (a). Reconstruction using a free latissimus dorsi muscle flap with split-thickness skin graft had been performed (b).

Patients with other malignancies such as hemangiopericy- or metastasis. At 1 month postoperatively, complete wound toma, Kaposi’s sarcoma, and malignant fibrous histiocytoma healing was verified, and then radiation therapy was initiated were also excluded. The risk factors of demographic data selectively based on the protocol of our institution. of patients, horizontal and vertical dimensions of resection, and adjuvant treatments were explored, and their effects on 2.3. Statistical Analysis. The Kaplan-Meier method was used the recurrence rates, metastasis, and patient survival were to determine overall survival, median survival, and disease- analyzed. free survival rates. To evaluate the association between risk factors and outcome parameters, the Cox regression test was 2.2. Treatment and Evaluations. Preoperative evaluation chosen for univariate and multivariate analyses. Statistical comprised pathologic confirmation of the skin biopsy spec- analysis was performed using version 9.3 of the SAS program imens or reconfirmation using histologic slides at other (SAS Institute Inc., Cary, North Carolina, USA). A 𝑃 value centers, magnetic resonance imaging (MRI) of the brain, <0.05 was considered statistically significant. chest radiography, computed tomography (CT) of the neck, positron emission tomography (PET), and bone scan of 3. Results the whole body. They all showed no evidence of metastasis to distant organs. Scalp hair was removed preoperatively, We retrospectively reviewed the records of a total of 14 and the lesion was reevaluated. For skin margins without patients (13 males and 1 female). The patients’ demographic definitive presentation, mapping biopsies were performed in characteristics are presented in Table 1. All patients were outpatient clinic to determine the extent of resection. In the Korean,andthemeanageatthediagnosiswas69(range:52– operating theater, the boundary of the primary lesion was 81) years, whereas the mean follow-up period was 17 (range: reconfirmed, and the lateral resection margins—free from 8–87) months. At the time of the initial diagnosis, 9 patients the tumor—were determined. Thereafter, a wide excision was had high blood pressure requiring antihypertensive medica- performed, while ensuring sufficient safety margins, to obtain tion, 1 patient had chronic obstructive pulmonary disease, 1 clear lateral and deep resection margins according to the patient had chronic kidney disease, and 1 patient had a history initial surgical plan (Table 1). After removing the primary of a cerebrovascular event. With regard to the postoperative lesion, an intraoperative frozen biopsy was performed to adjuvant treatment, 3 patients received radiotherapy alone, 3 establish clear surgical margins (Figure 1). The resultant patients received chemotherapy alone, and 5 patients received defect was reconstructed by the free latissimus dorsi (LD) both radiotherapy and chemotherapy. No revision surgery muscle flap. After successful microanastomosis between the due to vascular compromise of the flap was required in vascular pedicles of the flap (thoracodorsal (TD) artery and any of these cases. Moreover, postoperative complications vein) and the recipient vessels (superficial temporal (ST) such as hematoma, wound infection, or flap failure were not artery and vein), the raw surface was covered with the muscle observed. After radiation therapy was administered, partial flap and a split-thickness skin graft (Figure 1). Cervical lymph loss of the skin graft was observed in some cases. However, node dissection was performed only in patients who were sus- the degree of loss was not significant, and wound healing was pected to have regional lymph node metastasis, as confirmed completed by secondary intention. In 1 patient, an additional by preoperative imaging studies. Outpatient clinic-based split-thickness skin graft was applied to cover the raw surface follow-ups were performed postoperatively, every 3 months in the operating room. The overall 5- and 2-year survival rates during the first year and then every 6 months over the next were 15% and 75%, respectively, whereas the 5- and 2-year 1 year. During each visit, brain MRI and chest radiography disease-free survival rates were 7.7% and 38.7%, respectively. were performed in each patient for determining recurrence The mean survival duration was 32 months (Figures 2 and 3). BioMed Research International 3 Death (months) area Meta (months) Metastasis Local (months) recurrence local flap; SG: skingraft; MP: positive;margin MN: margin Profile of radiotherapy therapy Adjuvant results Biopsy methods Reconstruction emale; NR: not recorded; G: galea; P: periosteum; B: bone; FF: free flap;LF: Deep margin Table 1: Overall information of 14 patients with angiosarcoma of the scalp. NR B FF, SG MP C Null 11 FL Null FL Resection margin (cm) 5 2 B FF, SG MN RT, C 63 Gy 15 21 LRR 30 8 7 8 3 P FF, SG MN RT, C 50 Gy 11 NE Null NE 6.0 3 B FF, SG MN RT 42 Gy 5 NE Null NE 3 3 P FF, SG MN C Null 10 12 Brain 23 54 53 53 B6 1 B 2 FF, SG FF, SG 3 G P MN MN null B LF FF, SG null Null FF, SG MN MN Null RT, C null MN NE 17 61 Gy RT Null 45 50 Gy 36 8 Lung 7 Brain NE 48 10 9 37 NE Lung Lung Null NE 13 NE 2 1 G FF, SG MN C Null 5 69 Lung 85 5F,GNTC1y 13BoneNE 45PFF,SGMNRT,C61Gy9 5.5 2 P FF, SG MN RT NR 7 FL LRR FL 2.5 3 B FF, SG MN RT, C 30 Gy 13 20 Lung 32 × × × × × × × × × × × × × × × 9.5 8.5 at Dx (cm) Tumor size 81 11.5 81 3 71 5 61 4 52 3 75 75 13 76 5 58 3 70 4 70 8.5 68 6 64 9 64 3 Dx Age at F M M M M M M M M M M M M M Gender 5 6 7 8 11 12 13 14 9 10 Patient number 1 2 3 4 Dx: diagnosis; cm: centimeters; Meta area: area of metastasis; M: male; F: f negative; RT: radiotherapy; C: chemotherapy; Gy: grays; NE: no event; FL: follow-up loss; LRR: locoregional recurrence. 4 BioMed Research International

Survival function Survival function 1.0 1.0

0.8 0.8

0.6 0.6

0.4 0.4 Cumulative survival Cumulative Cumulative survival Cumulative

0.2 0.2

0.0 0.0

0.0 20.0 40.0 60.0 80.0 100.0 0.00 20.00 40.00 60.00 80.00 100.00 Month Month Survival function Survival function Censored Censored Figure 2: A Kaplan-Meier survival plot for overall survival rate. The Figure 3: A Kaplan-Meier survival plot for disease-free survival overall 5- and 2-year survival rates were 15% and 75%, respectively. rate. The 5- and 2-year disease-free survival rates were 7.7% and 38.7%, respectively.

Survival functions according to DM status Metastatic tumor dissemination to the lung or brain was associated with a major cause of death. Among the 14 1.0 patients, 7 patients were observed until the time of death. The major causes of death were the following: 4 cases of serious pulmonary complications as a result of lung metastasis and 0.8 2 cases of intracranial hemorrhage as a result of a metastatic braintumor.Thecommonchiefcomplaintswhenvisitingthe emergency room were sudden onset dyspnea or hemoptysis 0.6 in patients who had pulmonary metastasis and sudden onset altered mentality, limb weakness, or dysarthria in patients who had brain metastasis. None of the risk factors, except for 0.4 a deep surgical margin for excision, were significantly asso- Cumulative survival Cumulative ciated with the rates of recurrence, metastasis, and patient survival. Although a deep excision margin was the only factor 0.2 significantly associated with the rate of recurrence, it was not significantly associated with the rate of metastasis or thepatientsurvival.Thehazardratioforlocalrecurrence 0.0 was increased to 24.15 (95% confidence interval: 2.12–275.24) when the periosteum was preserved, as compared to when the 0.0 5.0 10.0 15.0 20.0 periosteum and bone were resected (𝑃 < 0.05) (Figure 4). Of Month the 14 patients, 1 experienced postoperative depression and was therefore referred to a psychiatrist. DM status B (bone) G (galea) P (periosteum)

4. Discussion Figure 4: Comparison of the recurrence rates based on the status of deep resection margins. Kaplan-Meier plots stratified by the status The poor prognosis of AS is reportedly associated with of deep resection margins are shown. Only a deep excision margin various factors. First, the definite diagnosis in these cases is was significantly related to the recurrence rate. The hazard ratio for often delayed or incorrect; therefore, the disease is already at local recurrence was increased to 24.15 (95% CI: 2.12–275.24) when an advanced stage during the initial presentation. AS usually the periosteum was preserved, as compared to when the periosteum has an insidious growth pattern, and the clinical features andbonewereresected(𝑃 value <0.05). CI: confidence interval. BioMed Research International 5

(a) (b) (c)

Figure 5: Various preoperative clinical features of angiosarcoma of the scalp: ulcerative type (a), bruise-like type (b), and nodular type (c).

may vary [15, 16] (Figure 5). Benign diseases such as heman- surgeons can acquire a more sufficient resection margin, free giomaorvascularmalformation,pigmentedskinlesions, fromthetumor,whenperformingwideexcisionofASinthe seborrheic dermatitis or seborrheic keratosis, nodular or scalp. ulcerated skin lesions, and inflammatory skin conditions may After the radical resection of AS, the wide defect was make differential diagnosis difficult [4]. Moreover, malignant reconstructed using free LD muscle flap and a split-thickness neoplasms, such as malignant melanoma, can mimic the skin graft. Risk factor analysis indicated that a deep margin presentation of AS [1]. These factors may often lead to the for excision was significantly associated with the rate of lackofrecognitionbypatientsaswellasthelackofperception recurrence. The periosteum should be stripped off to reduce by physicians. Hence, a high grade of suspicion and adequate the risk of local recurrence. A wide tissue defect without skin biopsy in the earlier stage of the disease are necessary any periosteum is recommended to be reconstructed using for accurate diagnosis. Second, vascular origin of AS may be free flap transfer, and the LD muscle flap may be a good associated with poor prognosis, and AS has a high incidence candidate regarding its size. In addition, TD vascular pedicles of hematogenous spread [2]. The most common distant have compatible calibers and sufficient length as compared to metastatic site for AS is the lung [4, 17], although the brain is ST recipient vessels. If a muscle flap is located under the skin also prone to the dissemination of tumor cells [17]. Third, AS graft, more rapid secondary healing is expected. The aesthetic tends to develop as multiple lesions, which makes it difficult outcome after reconstruction with the combination of free to define a clear boundary for the tumor [1]. In the cases of muscle flap and skin graft was acceptable, with considering satellite lesions, the residual tumor may be occult, even after scalpcontourandskincolormatching.Wedidnotobserve wide excision. Fourth, cutaneous AS is a very rare disease any complications related to the flap operation, such as even among malignant cancers, and the accumulated clinical hematoma under the flap, infection, or loss of the flap. experience on AS remains insufficient. Consequently, there is The 5-year survival rate in the present study was 15%, no clear consensus regarding the treatment protocol. which is consistent with the values reported in the litera- The anatomy of the scalp is unique and affects the clinical ture (10–50%) (Table 2) [4, 7, 11–13]. Complete excision is course of AS [16]. There is considerable arterial blood supply considered to be important for overall survival in patients to the scalp [18]; in particular, diverse branches originating with AS. However, only a few studies have described the from both internal and external carotid arterial systems anas- specific surgical protocol and its relationship with prognosis. tomose in the scalp. Compared to such horizontal patterns of The extent of the lateral safety margin was not significantly arterial supply, venous drainage displays both horizontal and correlated with outcome parameters, due to a small sample vertical patterns. The horizontal drainage of the superficial size. A larger number of patients in a future study may vein travels along the arterial pathway. However, there are indicate a statistical association between the extent of the also special vertical venous channels that are directly linked lateralsafetymarginandtheclinicaloutcomes.Thedepth with the intracranial space. These are termed emissary veins; of excision should reach the periosteum, in order to prevent due to the direct interconnection of these veins, the tumor tumor recurrence. caneasilyspreadintothebrainparenchyma.Thisanatomic The current primary treatment of AS involves complete characteristic may explain higher occurrences of brain metas- excision of the tumor and adjuvant radiotherapy; however, tasis; in fact, 2 of 7 patients in the present study died due the prognosis of AS remains obscure. We assume the mapping to intracranial hemorrhage after brain metastasis. Unlike the biopsy helps in determining surgical margin in advance, face, only a few important structures need to be preserved in and the effectiveness of such procedure has been affirmed the scalp, in terms of functional or aesthetic aspects. Hence, through previous studies in AS and extramammary Paget’s 6 BioMed Research International 31 MS 36.7 64.0 (5–76) (months) :and/or;STSG: ± (CI 27.0–37.0) (CI 48.4–75.6) (%) DFSR 2-year se-free survival rate; MS: medial (%) OSR 2-year (%) DFSR 5-year 15 7.7 75 38.7 12 NR 29 NR 15 12 NR NR NR 28.4 38 16 NR NR 25.2 54 NR 71 NR NR NR NR NR 36.2 NR NRNR NR NR 26 NR NR NR NR NR NR 22.1 NR 13.4 (%) 21.7 NR NR NR NR OSR 5-year e calvarium; Pr: ipsilateral parotidectomy; Nd: ipsilateral neck CT: chemotherapy; IT: immunotherapy; +: and; 13 17 32 24 NR NR NR 100 15 NR 19.7 13.7 18.3 55.3 29.5 25.2 (1–96) (8–87) period (3–159) (5–120) (5–108) (12–144) Mean f/u (3.2–106) (months) (SD 74.4) (4.7–227.1) (2.5–105.9) NR NR 33 20 NR NR NR NR NR NR 42 NR NR NR 20 WE WE STSG NR 29 NR NR NR 22 WE, LF (2), and method Pr + Nd Surgical (1–5 cm) (3–5 cm) +STSG+ STSG (26) +Oc+LD WE (5 cm) WE (2 cm) ± CT ± CT (7) CTOc+STSGNRNRNRNRNR17.2 IT (17), S ± ± ± CT (39) NR CT (11) ± RT (8) (6) (4) RT CT ± ± ± ± RT Treatment (1), CT (9) RT ± +RT+CT(2) S CT (6), RT + CT (1), and C (2) S + RT (3), RT (69) NR IT (20), RT (5), and IT ± S + RT (6), RT (3), S S(32),S+RT(38),S+ S (5), S + RT (5), RT (3) NR + CT + IT (2), RT S (1), S + RT (2), S + RT S (3), S + RT (3), S + CT S+RT S(6),S+RT(2),S+CT (3), and S + RT + CT (5) ± CT (4), and S + RT + CT S (12), S + CT (3), S + RT S+RT(19),S+CT(2),S Table 2: A review of literatures. face face face face neck neck neck neck neck neck Scalp S + RT (28), RT (1) of tumor Location Scalp and Scalp and Scalp and Scalp and Head and Head and Head and Head and Head and Head and C C NR NR NR Scalp S and Race A (1) A (1) A(J) Scalp RT+IT NR A(J) A(K) Scalp A(K) Scalp S C(4), C(19), C(28), AA (2) AA (3), confidence interval. % 71 71 71 71 74 63 77 77 67 C (17) 69 66 66 66 Dx NR NR Age at (52–81) (51–80) (12–75) (49–91) (52–83) (33–90) (49–83) (56–92) (59–86) (58–94) (64–82) (SD 14.4) 61 (6–83) NR 5 5 7 8 11 11 13 13 21 18 14 M 39 29 50 44 (1) (1) (5) (5) (7) (9) (6) (6) (4) (F) (11) (12) (19) (16) (28) (30) 9 6 13 18 19 14 14 𝑁 55 72 23 28 29 20 80 48 : number of patients; M: number of male patients; F: number of female patients; Dx: diagnosis; f/u: follow-up; OSR: overall survival rate; DFSR: disea 𝑁 Year 2015 2015 1987 2015 1995 2012 1979 2014 2010 1994 1996 2003 2005 2008 2008 [7] [6] [4] [11] [21] [14] [23] [20] al. [19] Authors ¨ ohler et al. et al. [12] et al. [16] et al. [24] Patel et al. Our study Mark et al. Dettenborn Morrison et Buschmann Pawlik et al. Mullins and Ogawa et al. Lydiatt et al. K Hodgkinson Ohguri et al. Holden et al. Hackman [5] Limetal.[22] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 survival; SD: standard deviation; NR:split-thickness not skin recorded; graft; C: WE: Caucasian; wide AA:dissection;cm:resectionmarginincentimeters;CI:95 excision; African American; LF: local A: Asian;flap; K: reconstruction LD: Korean; J: Japanese; using S:latissimus surgery; a free RT: dorsi radiation; muscle flap;Oc: outer corticotomy of th Year: year of publication; BioMed Research International 7 disease [25, 26]. This method facilitates the detection of [3] R. Sasaki, T. Soejima, K. Kishi et al., “Angiosarcoma treated occult satellite lesions, and it provides indications of the more with radiotherapy: impact of tumor type and size on outcome,” extensive resection. Aggressive excision of the tumor of the International Journal of Radiation Oncology Biology Physics,vol. scalp may be more beneficial for patient survival. Gudewer 52,no.4,pp.1032–1040,2002. et al. presented a case of whole scalp reconstruction with [4] C. A. Holden, M. F. Spittle, and E. W. Jones, “Angiosarcoma of afreemuscleflapandskingraftafterASresection,witha the face and scalp, prognosis and treatment,” Cancer,vol.59,no. good postoperative result [1]; the aesthetic results were also 5, pp. 1046–1057, 1987. satisfactory in that case. According to Lim et al. [22], radical [5] B. Mullins and T. Hackman, “Angiosarcoma of the head and resection with a sufficient safety margin (5 cm) may pro- neck,” International Archives of Otorhinolaryngology,vol.19,no. vide survival advantage. Up-to-date nonsurgical treatment 3, pp. 191–195, 2015. modalities such as immunotherapy or gene therapy may be [6]W.M.Lydiatt,A.R.Shaha,andJ.P.Shah,“Angiosarcomaofthe promising [5, 15, 20]. There are still ongoing debates about the head and neck,” The American Journal of Surgery,vol.168,no.5, application of chemotherapy in the standard AS treatment. pp.451–454,1994. Based on the current updates of literatures, chemotherapy is [7]R.J.Mark,J.C.Poen,L.M.Tran,Y.S.Fu,andG.F.Juillard, more recommended in case of recurring AS for the palliative “Angiosarcoma. A report of 67 patients and a review of the purpose[2,27–30].Weusedchemotherapyancillaryto literature,” Cancer,vol.77,no.11,pp.2400–2406,1996. surgical resections and radiotherapy. Further research and [8] P. P. Sordillo, R. Chapman, S. I. Hajdu, G. B. Magill, and R. B. clinical experience are necessary in order to apply new ther- Golbey, “Lymphangiosarcoma,” Cancer,vol.48,no.7,pp.1674– apeutic modalities. We suggest multidisciplinary approaches 1679, 1981. to AS patient with active discussions and cooperation among [9]K.T.K.Chen,K.D.Hoffman,andE.J.Hendricks,“Angiosar- physicians in different fields for maximizing the effect of coma following therapeutic irradiation,” Cancer,vol.44,no.6, treatments. For example, our institution has a sarcoma center pp. 2044–2048, 1979. in the cancer hospital. Medical collaborations were made [10] J. C. Maddox and H. L. Evans, “Angiosarcoma of skin and soft among departments of plastic and reconstructive surgery, tissue: a study of forty-four cases,” Cancer,vol.48,no.8,pp. dermatology, medical oncology, radiation oncology, and 1907–1921, 1981. neurosurgery, and feedback was exchanged before and after [11] T. M. Pawlik, A. F. Paulino, C. J. McGinn et al., “Cutaneous the surgery. angiosarcoma of the scalp: a multidisciplinary approach,” Can- The current study had some limitations, including the cer, vol. 98, no. 8, pp. 1716–1726, 2003. retrospective study design and small sample size. Moreover, [12]D.J.Hodgkinson,E.H.Soule,andJ.E.Woods,“Cutaneous no control group was used for comparison. Hence, a future angiosarcoma of the head and neck,” Cancer,vol.44,no.3,pp. studywithalargersamplesizemayprovidemoresignificant 1106–1113, 1979. information. [13]B.A.Guadagnolo,G.K.Zagars,D.Araujo,V.Ravi,T.D. Shellenberger,andE.M.Sturgis,“Outcomesafterdefinitive treatment for cutaneous angiosarcoma of the face and scalp,” 5. Conclusions Head & Neck,vol.33,no.5,pp.661–667,2011. [14] S. H. Patel, R. E. Hayden, M. L. Hinni et al., “Angiosarcoma AS is a rare and highly aggressive malignant tumor with a of the scalp and face: the Mayo Clinic experience,” JAMA poor prognosis. We observed that only a deep margin for Otolaryngology—Head & Neck Surgery,vol.141,no.4,pp.335– excision was significantly related to the rate of recurrence. 340, 2015. Hence, ensuring the presence of a sufficient deep margin as [15] S. Singla, P. Papavasiliou, B. Powers et al., “Challenges in the well as an adequate lateral margin is essential during wide treatment of angiosarcoma: a single institution experience,” The excision of AS. Further study is also needed to develop new American Journal of Surgery,vol.208,no.2,pp.254–259,2014. treatments for AS, such as immunotherapy or gene therapy. [16] A. Buschmann, M. Lehnhardt, N. Toman, P. Preiler, M. S. Salakdeh, and T. Muehlberger, “Surgical treatment of angiosar- Conflict of Interests coma of the scalp: less is more,” Annals of Plastic Surgery,vol. 61, no. 4, pp. 399–403, 2008. The authors declare that there is no conflict of interests [17]N.Naka,M.Ohsawa,Y.Tomita,H.Kanno,A.Uchida,andK. regarding the publication of this paper. Aozasa, “Angiosarcoma in Japan.A review of 99 cases,” Cancer, vol.75,no.4,pp.989–996,1995. References [18]D.E.Tolhurst,M.H.Carstens,R.J.Greco,andD.J.Hurwitz, “The surgical anatomy of the scalp,” Plastic and Reconstructive [1] E. Gudewer, E. Holzle,¨ and L. Li, “Widespread cutaneous Surgery,vol.87,no.4,pp.603–614,1991. angiosarcoma of the scalp: diagnosis and soft tissue recon- [19]W.H.Morrison,R.M.Byers,A.S.Garden,H.L.Evans,K.K. struction with a combined double-muscle-free flap and split- Ang, and L. J. Peters, “Cutaneous angiosarcoma of the head and thickness-skin graft,” OralandMaxillofacialSurgery,vol.13,no. neck. A therapeutic dilemma,” Cancer,vol.76,no.2,pp.319–327, 2, pp. 95–98, 2009. 1995. [2]W.M.Mendenhall,C.M.Mendenhall,J.W.Werning,J. [20]T.Ohguri,H.Imada,S.Nomotoetal.,“Angiosarcomaof D. Reith, and N. P. Mendenhall, “Cutaneous angiosarcoma,” the scalp treated with curative radiotherapy plus recombinant American Journal of Clinical Oncology: Cancer Clinical Trials, interleukin-2 immunotherapy,” International Journal of Radia- vol. 29, no. 5, pp. 524–528, 2006. tion Oncology Biology Physics,vol.61,no.5,pp.1446–1453,2005. 8 BioMed Research International

[21] H. F. Kohler,¨ R. I. Neves, E. R. Brechtbuhl,¨ N. V. Mattos Granja, M. K. Ikeda, and L. P. Kowalski, “Cutaneous angiosarcoma of the head and neck: report of 23 cases from a single institution,” Otolaryngology: Head and Neck Surgery,vol.139,no.4,pp.519– 524, 2008. [22] S.-Y.Lim, J.-K. Pyon, G.-H. Mun, S.-I. Bang, and K.-S. Oh, “Sur- gical treatment of angiosarcoma of the scalp with superficial parotidectomy,” Annals of Plastic Surgery,vol.64,no.2,pp.180– 182, 2010. [23] K. Ogawa, K. Takahashi, Y. Asato et al., “Treatment and prognosis of angiosarcoma of the scalp and face: a retrospective analysis of 48 patients,” The British Journal of Radiology,vol.85, no. 1019, pp. e1127–e1133, 2012. [24] T. Dettenborn, K. Wermker, H.-J. Schulze, M. Klein, V.Schwip- per, and C. Hallermann, “Prognostic features in angiosarcoma of the head and neck: a retrospective monocenter study,” Journal of Cranio-Maxillofacial Surgery,vol.42,no.8,pp.1623–1628, 2014. [25]R.Bullen,P.O.Larson,A.E.Landecketal.,“Angiosarcoma of the head and neck managed by a combination of multiple biopsies to determine tumor margin and radiation therapy. Report of three cases and review of the literature,” Dermatologic Surgery, vol. 24, no. 10, pp. 1105–1110, 1998. [26] B. J. Kim, S. K. Park, and H. Chang, “The effectiveness of mapping biopsy in patients with extramammary Paget’s disease,” Archives of Plastic Surgery,vol.41,no.6,pp.753–758,2014. [27]M.G.Fury,C.R.Antonescu,K.J.VanZee,M.F.Brennan,and R.G.Maki,“A14-yearretrospectivereviewofangiosarcoma: clinical characteristics, prognostic factors, and treatment outcomes with surgery and chemotherapy,” Cancer Journal, vol. 11, no. 3, pp. 241–247, 2005. [28] R. Isogai, A. Kawada, Y. Aragane, and T. Tezuka, “Successful treatment of pulmonary metastasis and local recurrence of angiosarcoma with docetaxel,” The Journal of Dermatology,vol. 31, no. 4, pp. 335–341, 2004. [29] K. M. Skubitz and P. A. Haddad, “Paclitaxel and pegylated- liposomal doxorubicin are both active in angiosarcoma,” Can- cer,vol.104,no.2,pp.361–366,2005. [30] U. Wollina, J. Fuller,¨ T. Graefe, M. Kaatz, and E. Lopatta, “Angiosarcoma of the scalp: treatment with liposomal doxorubicin and radiotherapy,” Journal of Cancer Research and Clinical Oncology,vol.127,no.6,pp.396–399,2001. Hindawi Publishing Corporation BioMed Research International Volume 2015, Article ID 579282, 8 pages http://dx.doi.org/10.1155/2015/579282

Research Article Computer-Aided Decision Support for Melanoma Detection Applied on Melanocytic and Nonmelanocytic Skin Lesions: A Comparison of Two Systems Based on Automatic Analysis of Dermoscopic Images

Kajsa Møllersen,1 Herbert Kirchesch,2 Maciel Zortea,3 Thomas R. Schopf,1 Kristian Hindberg,3 and Fred Godtliebsen3

1 Norwegian Centre for Integrated Care and Telemedicine, University Hospital of North Norway, 9038 Tromsø, Norway 2PrivateOffice,VenloerStraße107,50259Pulheim,Germany 3Department of Mathematics and Statistics, UiT The Arctic University of Norway, 9037 Tromsø, Norway

Correspondence should be addressed to Kajsa Møllersen; [email protected]

Received 11 September 2015; Accepted 3 November 2015

Academic Editor: Elisabeth Roider

Copyright © 2015 Kajsa Møllersen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Commercially available clinical decision support systems (CDSSs) for skin cancer have been designed for the detection of melanoma only. Correct use of the systems requires expert knowledge, hampering their utility for nonexperts. Furthermore, there are no systems to detect other common skin cancer types, that is, nonmelanoma skin cancer (NMSC). As early diagnosis of skin cancer is essential, there is a need for a CDSS that is applicable to all types of skin lesions and is suitable for nonexperts. Nevus Doctor (ND) is a CDSS being developed by the authors. We here investigate ND’s ability to detect both melanoma and NMSC and the opportunities for improvement. An independent test set of dermoscopic images of 870 skin lesions, including 44 melanomas and 101 NMSCs, were analysed by ND. Its sensitivity to melanoma and NMSC was compared to that of Mole Expert (ME), a commercially available CDSS, using the same set of lesions. ND and ME had similar sensitivity to melanoma. For ND at 95% melanoma sensitivity, the NMSC sensitivity was 100%, and the specificity was 12%. The melanomas misclassified by ND at 95% sensitivity were correctly classified by ME, and vice versa. ND is able to detect NMSC without sacrificing melanoma sensitivity.

1. Introduction The dermoscope (dermatoscope, epiluminescence micro- scope) reveals structures not visible to the naked eye and Melanoma is the deadliest of all skin cancers. When it is has proven to raise diagnostic accuracy of melanoma when detected early, the treatment is excision of the tumour, and used by properly trained personnel [5]. Dermoscopy has the survival rate is high. Recent developments in melanoma not reached widespread use among general practitioners treatment are promising [1, 2], but the survival rate for (GPs), except in Australia [6]. In view of increased melanoma patients with metastasised melanoma is still poor [3, 4]. incidence rates, national screening programmes, and growing Early diagnosis is crucial, but challenging, since early stage awareness of skin cancer in the public, there seems to be a melanomas resemble benign skin lesions. Other types of need for clinical decision support systems (CDSSs) for GPs skin cancer, like basal cell carcinoma (BCC) and squamous not familiar with dermoscopy. cell carcinoma (SCC), have high incidence rates but low For the past few decades, many CDSSs (also referred mortality rates [4]. Early detection is beneficiary for the to as computer-aided diagnostic (CAD) systems) have been patient to get early treatment and avoid further damaging of developed for melanoma detection. For a description of the the surrounding skin. major steps and an overview of different technologies, see, 2 BioMed Research International for example, [7]. The review paper of Rosado et al. [8] from 2. Materials and Methods 2003 concluded that the diagnostic accuracy of the CDSSs was statistically not superior to that of human diagnosis. 2.1. Data. From March to December 2013, patients were Vestergaard and Menzies [9] reported the same in 2008, and recruited at a private dermatology practice in Pulheim, Korotkov and Garcia [10] made a similar conclusion in 2012. Germany. Adult patients scheduled for excision of a pig- Simultaneously high sensitivity and specificity scores have mented skin lesion were eligible for inclusion. Further- been reported [11], but the scores dropped when the same sys- more, patients who were having nonpigmented skin lesions tems were tested in clinical-like settings with an independent excised were eligible for inclusion if melanoma, BCC, or test set of consecutively collected images. Dreiseitl et al. [12] SCC was a potential differential diagnosis. Patients attending concluded that the performance of their system in a clinical- theGermanskincancerscreeningprogramme[35]were like setting was significantly lower than the result obtained also eligible for inclusion if they had skin lesions selected for excision. Informed written consent was obtained from during training. Elbaum et al. [13] reported 100% sensitivity all patients prior to inclusion. Skin lesions were excised and 85% specificity on the training set, with a dramatic drop becauseofconcernaboutmalignancyorwhenrequestedby in specificity to 9% on an independent test set [14]. Also, the patient for other reasons. All skin lesions were photo- Bauer et al. reported better results for the training set [15] than graphed prior to excision with a digital camera (Canon G10, those achieved with an independent test set [16]. The good Canon Inc., Tokyo, Japan) with an attached dermoscope results of Hoffmann et al. [17] were reproduced in a small (DermLite FOTO, 3Gen LLC, California, USA) and with a study with only 6 melanomas [18] but declined in another videodermoscope (DermoGenius ultra, DermoScan GmbH, small study [19]. In two other small studies with independent Regensburg, Germany). All excised lesions were examined by test sets [20, 21], the performance reported by Blum et al. [22] a dermatopathologist. In the case of a malignant diagnosis, a declined. There are several possible explanations for the drop second dermatopathologist examined the excised lesion and in performance. In studies where cross-validation has been a consensus diagnosis was set. used to validate the performance, the whole data set is used for feature selection or model selection, which gives overly 2.2. Automatic Image Analysis. ND takes a dermoscopic optimistic results [23, 24]. Also, several studies exclude non- image from the Canon/DermLite device as input and clas- melanocytic lesions post hoc based on the pathology reports, sifies the lesion. ND is still in an experimental phase. Ina which introduces bias. previous study, ND performed as well as three independent Differentiating between melanocytic and nonmelano- dermatologists in terms of melanoma sensitivity and speci- cytic lesions can be challenging even for experienced dermo- ficity [36]. In another study, ND performed as well asan scopy users [25], and it is recommended that a CDSS for independent dermatologist on an independent test set con- melanoma detection can handle nonmelanocytic lesions as sisting of 21 melanomas and 188 benign lesions [37]. The data well [8, 11, 26]. A lesion is classified as suspicious if it resem- set in this study partly overlaps with the test set in Møllersen bles a melanoma, but lesions that resemble NMSC should also et al. [37], but not with the training set. ND has not been be classified as suspicious, especially when used by GPs [27], retrained, so the present data set is independent. ND outputs and not be classified together with nonsuspicious lesions. a probability of malignancy for each lesion image, and the sensitivity can be tuned with a parameter 𝛼 (see [36–38] for In this report we present the performance of a CDSS for details). melanoma detection, Nevus Doctor (ND), when applied to both melanocytic and nonmelanocytic lesions. To our knowl- ME (MoleExpert micro Version 3.3.30.156) takes a der- edge, this has not been previously presented for image-based moscopic image from the DermoGenius device as input. The output is a number between −5.00 and 5.00,wherehigh CDSSs. Shimizu et al. [28] included BCC but excluded other values indicate suspicion of melanoma, and the sensitivity NMSCs. Several studies have included nonmelanocytic can be tuned by adjusting a threshold 𝑡. ME is intended lesions, but without reporting sensitivity to NMSC [14–17, for use on melanocytic lesions only. ME was chosen as the 29–31] or with less than three NMSCs [20, 32]. Recently, comparison system due to availability and the fact that it has non-image-based technologies for melanoma detection have been tested and compared to other commercial systems in included NMSC sensitivity in their reported findings [33, 34]. clinical settings on a small scale [29, 32]. To our knowledge, There is no apparent ranking of the CDSSs for melanoma the study of Perrinaud et al. [32] is the only study that detection. In practice, CDSSs can only be compared if tested compares several systems on the same set of lesions, and it on the same set of lesions, as done by Perrinaud et al. [32]. is therefore not possible to pick the most adequate reference We therefore compare the performance of ND to that of a system. commercially available system, Mole Expert (ME), on the same set of lesions. This methodology potentially identifies 2.3. Statistical Analysis. All excised lesions for which a the diagnostic difficulty of the data set, supplementing the pathology report was available were included in the analysis. information on the proportion of melanomas in situ, Breslow Cases were excluded if either of the CDSSs could not give depth, clinical diagnoses, and so forth. It also allows for an output or if images were missing. Presence of hairs and indirect comparison of CDSSs if, in the future, some study bubbles, lesion size, inadequate segmentation, and so forth chosetocompareanothersystemtoME. were not used as exclusion criteria. The data was divided BioMed Research International 3 into four classes according to the histopathological diag- Table 1: Histopathological diagnoses for the 877 skin lesions. nosis: melanoma, NMSC, benign melanocytic lesions, and Total: 877 benign nonmelanocytic lesions. The melanoma class includes Histopathological diagnosis 727 lesions where malignancy could not be ruled out by the Benign lesions dermatopathologists (ICD-10 D48.5), and the NMSC class Melanocytic lesions 596 includes precancerous lesions, as done in other studies [33, Naevus 574 34]. Precancerous lesions should be classified as suspicious, Naevoid lentigo 9 since the patient should receive treatment or follow-up. Blue naevus 8 The sensitivity and specificity scores of ND and ME were Spitz naevus 4 calculated by adjusting parameters 𝛼 and 𝑡 and classifying Sutton naevus 1 the lesions accordingly. Sensitivity refers to the ratio of 118 malignant lesions classified as suspicious to the total number Nonmelanocytic lesions of malignant lesions. Since there are two classes of malignant Seborrheic keratosis 80 lesions, the terms melanoma sensitivity and NMSC sensitivity Lentigo senilis 9 are used for clarification when needed. Neurofibroma 5 The clinical diagnoses are taken from the dermatologist’s Dermatofibroma 8 referrals to pathology. The clinical malignant class includes Hemangioma 4 lesions where malignancy is a differential diagnosis. The Verruca 2 clinical benign class consists of lesions that were given a Acanthoma 3 benign diagnosis by the dermatologist but were referred to pathology, which indicates that malignancy could not be Papilloma, fibroma molle, sebaceous gland hyperplasia, scar, eczema, folliculitis, and 7 ruled out by the dermatologist, even if this was not explicitly diskoid lupus erythematodes stated. Collision tumours 13 148 3. Results and Discussion Malignant and precancerous lesions Melanoma 45 3.1. Results. There were 516 consultations (47% women) Invasive melanoma 25 included in the study and a total of 877 excised lesions. Melanoma in situ 20 The minimum age was 18, the maximum age was 93, and Nonmelanoma skin cancers 103 median age was 53 years. Table 1 shows the histopathological Adnexal carcinoma 1 diagnoses. The ratio of benign lesions per melanoma was 16 : 1, which is within the range for dermatologists [39, 40]. Actinic keratosis 13 In total, 5% of the lesions were melanomas. The median Basalcellcarcinoma 71 Breslow depth for the 23 invasive melanomas was 0.50 mm, Squamous cell carcinoma 7 and the maximum Breslow depth was 2.25 mm. Table 2 Bowen’s disease 11 shows the diagnoses and the Breslow depths according to No histopathological diagnosis 2 the 2009 American Joint Committee on Cancer (AJCC) staging [41]. About 70% of the melanomas and about 90% of the NMSCs were clinically diagnosed as malignant. One Table 2: Characteristics of the 45 melanomas. lesion lacked clinical diagnosis. Of the 875 lesions with histopathological diagnosis, four were excluded because ME Diagnosis/Breslow Number did not give an output (one naevus, one seborrheic keratosis, In situ 13 one BCC, and one SCC) and one was excluded because the Lentigo maligna 7 Canon/DermLite image was lost (melanoma in situ), which Cutaneous metastasis 2 1 corresponds to less than %oftheimages.Incomparison, ≤1.00 mm 19 10 % of the lesions were excluded due to lesions size or device 1.01–2.00 mm 3 malfunctions in the study by Malvehy et al. [33]. 2.01–4.00 mm 1 Figure 1(a) shows receiver operating characteristic (ROC) >4.00 mm 0 curvesforNDandME.Theredandbluesolidcurvesshow sensitivity versus specificity for the whole data set where the malignant class includes both melanoma and NMSC, and a clear distinction between ND and ME can be seen. ND and ME, there is an overlap between melanomas in situ The pink and turquoise dotted curves show sensitivity versus and benign melanocytic lesions, whereas invasive melanomas specificity for melanocytic lesions only, and there is no can be separated from benign melanocytic lesions. ND has significant difference between ND and ME. Figure 1(b) shows high scores for NMSC, but also for benign nonmelanocytic NMSC sensitivity as a function of melanoma sensitivity. lesions. ME has greater relative variety for all categories than The boxplots in Figure 2 illustrate each CDSS’s ability to ND has which lead to larger overlaps and thus more difficulty discriminate between the different classes of lesions. For both in distinguishing the different categories. 4 BioMed Research International

1 1

0.9 0.9

0.8 0.8

0.7 0.7

0.6 0.6

0.5 0.5

Sensitivity 0.4 0.4 Sensitivity to NMSC to Sensitivity 0.3 0.3

0.2 0.2

0.1 0.1

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Specificity Sensitivity to melanoma

ND-all lesions ME-all lesions ND ND-melanocytic ME-melanocytic ME (a) (b)

Figure 1: (a) Receiver operating characteristic curves for Nevus Doctor and Mole Expert. (b) Nonmelanoma skin cancer sensitivity as a function of melanoma sensitivity.

1 5 0.8 3 0.6 1 0.4 −1 0.2 −3 0 −5 SCC (6) SCC (6) Other Other BCC (70) BCC (70) In situ In In situ In Sebker (79) Sebker Sebker (79) Sebker NMSC (25) NMSC (25) Invasive Invasive Other Other Other Other Naevus (573) Naevus Naevus (573) Naevus Collision (13) Collision Collision (13) Collision melanoma (19) melanoma (25) melanoma melanoma (19) melanoma (25) melanoma Melanocytic (22) Melanocytic Melanocytic (22) Melanocytic Nonmelanocytic (38) Nonmelanocytic Nonmelanocytic (38) Nonmelanocytic (a) (b)

Figure 2: Boxplots for (a) Nevus Doctor and (b) Mole Expert. The horizontal lines inside the boxes are the medians; the boxes are definedby the 25th and 75th percentiles. The whiskers are situated at two standard deviations from the mean. The crosses indicate outlier observations.

For ND at 95% melanoma sensitivity, the overall sensitiv- have shown that ND performed similarly to ME under ity was 99%, the specificity was 12%,thepositivepredictive the circumstances for which ME is intended. When non- value (PPV) was 18%, and the negative predictive value melanocytic lesions were included, ND performed better (NPV) was 98%, where sensitivity, specificity, PPV, and NPV than ME. Figure 1(b) shows that ND’s NMSC sensitivity are defined in terms of malignant (melanoma and NMSC) reached 100% at melanoma sensitivity of 86%, which means and benign histopathological diagnosis. Figure 3 shows the that, at reasonably high melanoma sensitivity, all NMSCs two melanomas that were misclassified at 95% melanoma were classified as suspicious. sensitivity for each CDSS. They were all in situ and were all The majority of the benign lesions were excised dueto clinically diagnosed as benign. suspicion of malignancy, and about 30% of the melanomas had a benign clinical diagnosis, so the overlap between 3.2. Discussion. The ROC curves for ND and ME in benign melanocytic lesions and melanomas in situ, as seen in Figure 1(a) show that ND performed similar to ME when Figure 2, was expected. ND misclassifies seborrheic keratoses nonmelanocytic lesions were excluded, and by that we as suspicious, similar to other CDSSs [20, 31–33]. GPs excise BioMed Research International 5

(a) (b) (c) (d)

(e) (f) (g) (h)

Figure 3: (a)–(d) Lesions photographed with Canon/DermLite. (e)–(h) Lesions photographed with DermoGenius. ((a)-(b) and (e)-(f)) The two melanomas misclassified by ND at 95% melanoma sensitivity. ((c)-(d) and (g)-(h)) The two melanomas misclassified by ME at 95% melanoma sensitivity.

1 only one dermatologist’s opinion, which is a drawback since 0.95 interobserver agreement is only moderate for dermatologists [25]. Short-term follow-up of the patient or consensus diag- 0.9 nosis based on dermoscopic and clinical images can be used

Sensitivity 0.85 as the gold standard for nonexcised lesions, but it requires large resources. The data set used in this study is independent 0.8 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 of all stages of ND’s development, but the lesions in this set Specificity are possibly more similar to the lesions ND has been trained on than to the lesions ME has been trained on, and this is ND, all lesions ME, all lesions potentially an advantage for ND. ND, melanocytic ME, melanocytic The present data set consisted of 44% melanomas in situ and median Breslow depth of 0.50 mm, which indicate Figure 4: Receiver operating characteristic curves for Nevus Doctor high diagnostic difficulty. A proportion of 31% melanomas and Mole Expert. Upper right part of Figure 1(a). in situ was reported on a similar population in Germany [43]. The data set consists of 44 melanomas, which is more than most reported studies of CDSSs with independent test more seborrheic keratoses per melanoma than dermatolo- sets, with some exceptions [14, 33]. With 44 melanomas, gistsdo[39,40,42],sothereisapotentialforsavingpathol- 95% melanoma sensitivity means that only two melanomas ogy resources if a CDSS can classify them correctly. Devel- are misclassified, and the results are therefore very sensitive opment of image analysis features to discriminate seborrheic to small changes in the data set. Confidence intervals for keratosis from melanoma has just begun [28]. highly data sensitive results can give a false impression of Only the segment of the ROC curve with high sensitivity generalisability and are therefore not reported. to melanoma, as shown in Figure 4, is clinically relevant, ND and ME misclassified different melanomas, as shown and therefore the area under the curve (AUC) is not an in Figure 3, which is not unexpected since the two CDSSs adequate summary of a CDSS’s performance. The area under have been developed independently of each other. It has been theclinicallyrelevantsegmentoftheROCcurvewouldbea reported that the sensitivity and specificity for dermatologists more adequate summary statistic; however, then the clinically improvewhenmajorityvoteorconsensusisusedtodiagnose relevant segment must first be defined. skin lesions [5, 25]. We have investigated whether a combi- Inadequate segmentation was not used as exclusion cri- nation of ND and ME will increase the sensitivity without terion, although it seems unlikely that a user would trust decreasing the specificity. Figure 5 shows the ROC curve for theoutcomeifthesegmentationfails.Butsincethereisno a classifier which classifies a lesion as suspicious if either ground truth for segmentation, it is less suited as exclusion ND or ME classifies it as suspicious. For melanocytic lesions criterion. The decision to carry out excision was based on only, this classifier outperformed ND and ME for sensitivities 6 BioMed Research International

1.00

0.95

0.9

Sensitivity 0.85

0.8 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0.00 Specificity

ND ND + ME ME

Figure 5: Receiver operating characteristic curves for Nevus Doctor, Mole Expert, and the combination classifier. Melanocytic lesions only.

(a) (b)

Figure 6: Lesion (a) was clinically diagnosed as naevus, whereas lesion (b) was clinically diagnosed as melanoma.

above 90%. For melanoma and NMSC, the performance was the system as shown in Figure 2 and also illustrated by about the same as for ND (not shown). An explanation of Malvehy et al. [33]. Whether the superior performance of the results can be that the lesions were photographed with Piccolo et al. [16] is due to few melanomas in situ is unknown, different cameras and dermoscopes, but it can also be because asthiswasnotreported.Thetwostudieswithmorethan of the different feature algorithms or different statistical 100 melanomas in an independent test set reported similar classifiers. sensitivities (98% and 97%) but different specificities (9% and A CDSS is not needed for skin lesions that obviously are 34%),butitisnotpossibletoconcludethatonesystemis melanomas, and it can therefore be argued that these lesions better than the other, since the exclusion criteria for the two should be excluded [14]. To evaluate ND’s performance studies are very different. A publicly available data set has under these conditions, all melanomas that were clinically been called for [8], but the wide variety of technologies is a diagnosed as melanoma with no benign differential diagnosis challenge. were excluded. There were then 22 melanomas left. Tun- ing the parameters to 95% melanoma sensitivity (21 of 22 melanomas detected) gave a specificity of 6% for ND and 13% 4. Conclusions for ME, similar to the 9% specificity reported by Monheit et al. [14]. One should, however, be very cautious when drawing We have shown that ND was able to detect NMSC without conclusions. To demonstrate the dependence on the data set, sacrificing melanoma sensitivity but misclassified benign one of the melanomas clinically diagnosed as benign was nonmelanocytic lesions. Although there are promising results replaced by a melanoma clinically diagnosed as malignant, for other technologies, dermoscopy is still the only widely shown in Figure 6. The specificity then increased to 12% for usedtoolforskinlesiondiagnosis.Nonmelanocyticlesions ND and decreased to 7% for ME, which is outside the range are an important aspect in melanoma detection, and more of the respective confidence intervals. research is needed, especially on features for differentiating Compared to other studies with independent test sets between melanomas and seborrheic keratoses. andaminimumof10melanomas[12,14,16,33,34],the Different inclusion and exclusion criteria, moderate-sized sensitivity and specificity scores of ND and ME are not data sets, and variety in the diagnostic difficulty make the superior. Inclusion of nonexcised lesions [12, 34] can have a reported sensitivity and specificity scores inadequate for positive effect on the observed specificity, since these lesions comparing different CDSSs. The demonstration of the results’ do not resemble malignant lesions. A higher proportion of dependence on the data set emphasises the need for direct melanomas in situ decreases the observed performance of comparison on the same set of lesions. Which system is the BioMed Research International 7

most adequate for comparison will remain unknown until American Academy of Dermatology,vol.44,no.2,pp.207–218, more studies are reported. 2001. According to a study by Dreiseitl and Binder [44], physi- [14] G. Monheit, A. B. Cognetta, L. Ferris et al., “The performance of cians are willing to follow the recommendation of a CDSS, MelaFind: a prospective multicenter study,” Archives of Derma- especially if they are not confident in their own diagnosis, and tology, vol. 147, no. 2, pp. 188–194, 2011. Fruhauf¨ et al. [20] reported that patients accept the use of a [15] P. Bauer, P. Cristofolini, S. Boi et al., “Digital epiluminescence CDSS. Hence, there is a potential for CDSSs in melanoma microscopy: usefulness in the differential diagnosis of cuta- detection if the systems can give reliable recommendations neous pigmentary lesions. A statistical comparison between for all kinds of lesions. visual and computer inspection,” Melanoma Research,vol.10, no. 4, pp. 345–349, 2000. Conflict of Interests [16]D.Piccolo,A.Ferrari,K.Peris,R.Daidone,B.Ruggeri,andS. Chimenti, “Dermoscopic diagnosis by a trained clinician vs. The authors declare that there is no conflict of interests a clinician with minimal dermoscopy training vs. computer- regarding the publication of this paper. aided diagnosis of 341 pigmented skin lesions: a comparative study,” British Journal of Dermatology,vol.147,no.3,pp.481– 486, 2002. References [17] K. Hoffmann, T. Gambichler, A. Rick et al., “Diagnostic and [1] C. Garbe, T. K. Eigentler, U. Keilholz, A. Hauschild, and J. M. neural analysis of skin cancer (DANAOS). A multicentre study Kirkwood, “Systematic review of medical treatment in mela- for collection and computer-aided analysis of data from pig- noma: current status and future prospects,” The Oncologist,vol. mented skin lesions using digital dermoscopy,” British Journal 16, no. 1, pp. 5–24, 2011. of Dermatology,vol.149,no.4,pp.801–809,2003. [2] A. Niezgoda, P. Niezgoda, and R. Czajkowski, “Novel [18] M. Barzegari, H. Ghaninezhad, P. Mansoori, A. Taheri, Z. approaches to treatment of advanced melanoma: a review on S. Naraghi, and M. Asgari, “Computer-aided dermoscopy for targeted therapy and immunotherapy,” BioMed Research Inter- diagnosis of melanoma,” BMC Dermatology,vol.5,no.1,article national,vol.2015,ArticleID851387,16pages,2015. 8, 2005. [3] American Cancer Society, “Cancer facts & figures 2014,” Tech. [19]J.C.Boldrick,C.J.Layton,J.Nguyen,andS.M.Swetter,“Eval- Rep., American Cancer Society, 2014. uation of digital dermoscopy in a pigmented lesion clinic: clini- [4] Cancer Registry of Norway, “Cancer in Norway 2013—cancer cian versus computer assessment of malignancy risk,” Journal of incidence, mortality, survival and prevalence in Norway,” Tech. the American Academy of Dermatology,vol.56,no.3,pp.417– Rep., Cancer Registry of Norway, 2015. 421, 2007. [5]H.Kittler,H.Pehamberger,K.Wolff,andM.Binder,“Diagnos- [20] J. Fruhauf,¨ B. Leinweber, R. Fink-Puches et al., “Patient accep- tic accuracy of dermoscopy,” The Lancet Oncology,vol.3,no.3, tance and diagnostic utility of automated digital image analysis pp.159–165,2002. of pigmented skin lesions,” Journal of the European Academy of [6]C.Rosendahl,G.Williams,D.Eleyetal.,“Theimpactofsub- Dermatology and Venereology,vol.26,no.3,pp.368–372,2012. specialization and dermatoscopy use on accuracy of melanoma diagnosis among primary care doctors in Australia,” Journal of [21] A. Fueyo-Casado, F. Vazquez-L´ opez,´ J. Sanchez-Martin, B. the American Academy of Dermatology,vol.67,no.5,pp.846– Garcia-Garcia, and N. Perez-Oliva,´ “Evaluation of a program for 852, 2012. the automatic dermoscopic diagnosis of melanoma in a general [7]A.MasoodandA.A.Al-Jumaily,“Computeraideddiagnostic dermatology setting,” Dermatologic Surgery,vol.35,no.2,pp. support system for skin cancer: a review of techniques and algo- 257–259, 2009. rithms,” International Journal of Biomedical Imaging,vol.2013, [22] A. Blum, R. Hofmann-Wellenhof, H. Luedtke et al., “Value of Article ID 323268, 2013. the clinical history for different users of dermoscopy compared [8] B. Rosado, S. Menzies, A. Harbauer et al., “Accuracy of computer with results of digital image analysis,” Journal of the European diagnosis of melanoma: a quantitative meta-analysis,” Archives Academy of Dermatology and Venereology,vol.18,no.6,pp. of Dermatology,vol.139,no.3,pp.361–367,2003. 665–669, 2004. [9] M. E. Vestergaard and S. W. Menzies, “Automated diagnostic [23]P.Smialowski,D.Frishman,andS.Kramer,“Pitfallsofsuper- instruments for cutaneous melanoma,” Seminars in Cutaneous vised feature selection,” Bioinformatics,vol.26,no.3,pp.440– Medicine and Surgery,vol.27,no.1,pp.32–36,2008. 443, 2009. [10] K. Korotkov and R. Garcia, “Computerized analysis of pig- [24] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statis- mented skin lesions: a review,” Artificial Intelligence in Medicine, tical Learning: Data mining, Inference, and Prediction, Springer vol.56,no.2,pp.69–90,2012. Series in Statistics, Springer, New York, NY, USA, 2nd edition, [11] A. Blum, I. Zalaudek, and G. Argenziano, “Digital image anal- 2009. ysis for diagnosis of skin tumors,” Seminars in Cutaneous Medi- cine and Surgery,vol.27,no.1,pp.11–15,2008. [25] G. Argenziano, H. P. Soyer, S. Chimenti et al., “Dermoscopy of [12] S. Dreiseitl, M. Binder, K. Hable, and H. Kittler, “Computer pigmented skin lesions: results of a consensus meeting via the versus human diagnosis of melanoma: evaluation of the feasibil- internet,” Journal of the American Academy of Dermatology,vol. ity of an automated diagnostic system in a prospective clinical 48,no.5,pp.679–693,2003. trial,” Melanoma Research,vol.19,no.3,pp.180–184,2009. [26] S. M. Rajpara, A. P. Botello, J. Townend, and A. D. Ormerod, [13] M. Elbaum, A. W. Kopf, H. S. Rabinovitz et al., “Automatic dif- “Systematic review of dermoscopy and digital dermoscopy/ ferentiation of melanoma from melanocytic nevi with multi- artificial intelligence for the diagnosis of melanoma,” British spectral digital dermoscopy: a feasibility study,” Journal of the Journal of Dermatology,vol.161,no.3,pp.591–604,2009. 8 BioMed Research International

[27] F. M. Walter, H. C. Morris, E. Humphrys et al., “Effect of adding [41] C. M. Balch, J. E. Gershenwald, S.-J. J. Soong et al., “Final version a diagnostic aid to best practice to manage suspicious pig- of 2009 AJCC melanoma staging and classification,” Journal of mented lesions in primary care: randomised controlled trial,” Clinical Oncology,vol.27,no.36,pp.6199–6206,2009. British Medical Journal,vol.345,no.7866,ArticleIDe4110,2012. [42] M. I. Duque, J. R. Jordan, A. B. Fleischer Jr. et al., “Frequency of [28] K. Shimizu, H. Iyatomi, M. E. Celebi, K.-A. Norton, and M. seborrheic keratosis biopsies in the United States: a benchmark Tanaka, “Four-class classification of skin lesions with task of skin lesion care quality and cost effectiveness,” Dermatologic decomposition strategy,” IEEE Transactions on Biomedical Engi- Surgery,vol.29,no.8,pp.796–801,2003. neering,vol.62,no.1,pp.274–283,2015. [43] A. Katalinic, A. Waldmann, M. A. Weinstock et al., “Does skin [29] M. J. Jamora, B. D. Wainwright, S. A. Meehan, and J.-C. Bystryn, cancer screening save lives?: an observational study comparing “Improved identification of potentially dangerous pigmented trends in melanoma mortality in regions with and without skin lesions by computerized image analysis,” Archives of Der- screening,” Cancer,vol.118,no.21,pp.5395–5402,2012. matology,vol.139,no.2,pp.195–198,2003. [44] S. Dreiseitl and M. Binder, “Do physicians value decision [30] M. Carrara, A. Bono, C. Bartoli et al., “Multispectral imaging support?Alookattheeffectofdecisionsupportsystemson and artificial neural network: mimicking the management physician opinion,” Artificial Intelligence in Medicine,vol.33,no. decision of the clinician facing pigmented skin lesions,” Physics 1,pp.25–30,2005. in Medicine and Biology,vol.52,no.9,pp.2599–2613,2007. [31] S. W. Menzies, L. Bischof, H. Talbot et al., “The performance of solarscan: an automated dermoscopy image analysis instrument for the diagnosis of primary melanoma,” Archives of Dermatol- ogy,vol.141,no.11,pp.1388–1396,2005. [32] A. Perrinaud, O. Gaide, L. E. French, J.-H. Saurat, A. A. Marghoob, and R. P.Braun, “Can automated dermoscopy image analysis instruments provide added benefit for the dermatologist? A study comparing the results of three systems,” British Journal of Dermatology,vol.157,no.5,pp.926–933,2007. [33] J. Malvehy, A. Hauschild, C. Curiel-Lewandrowski et al., “Clinical performance of the Nevisense system in cutaneous melanoma detection: an international, multicentre, prospective and blinded clinical trial on efficacy and safety,” The British Journal of Dermatology,vol.171,no.5,pp.1099–1107,2014. [34] H. Lui, J. Zhao, D. McLean, and H. Zeng, “Real-time Raman spectroscopy for in vivo skin cancer diagnosis,” Cancer Research,vol.72,no.10,pp.2491–2500,2012. [35] A. Waldmann, S. Nolte, M. A. Weinstock et al., “Skin cancer screening participation and impact on melanoma incidence in Germany—an observational study on incidence trends in regions with and without population-based screening,” British Journal of Cancer,vol.106,no.5,pp.970–974,2012. [36]M.Zortea,T.R.Schopf,K.Thonetal.,“Performanceofader- moscopy-based computer vision system for the diagnosis of pigmented skin lesions compared with visual evaluation by experienced dermatologists,” Artificial Intelligence in Medicine, vol.60,no.1,pp.13–26,2014. [37] K. Møllersen, M. Zortea, K. Hindberg, T. R. Schopf, S. O. Skrøvseth, and F. Godtliebsen, “Improved skin lesion diagnostics for general practice by computer aided diagnostics,”in Der- moscopy Image Analysis, Digital Imaging and Computer Vision, pp. 247–292, CRC Press, 2015. [38] M.Zortea,S.O.Skrøvseth,T.R.Schopf,H.M.Kirchesch,andF. Godtliebsen, “Automatic segmentation of dermoscopic images by iterative classification,” International Journal of Biomedical Imaging, vol. 2011, Article ID 972648, 19 pages, 2011. [39]R.Marks,D.Jolley,C.McCormack,andA.P.Dorevitch,“Who removes pigmented skin lesions? A study of the ratio of melanoma to other benign pigmented tumors removed by different categories of physicians in Australia in 1989 and 1994,” Journal of the American Academy of Dermatology,vol.36,no.5,pp.721– 726, 1997. [40] B. Lindelof,M.-A.Hedblad,andU.Ringborg,“Nevuseller¨ malignt melanom? Ratt¨ kompetens vid diagnostik ger lagre¨ kostnader,” Lakartidningen¨ ,vol.105,no.39,pp.2666–2669, 2008. Hindawi Publishing Corporation BioMed Research International Volume 2015, Article ID 642549, 9 pages http://dx.doi.org/10.1155/2015/642549

Research Article Wide Local Excision for Dermatofibrosarcoma Protuberans: A Single-Center Series of 90 Patients

Byung Jun Kim, Hyeonwoo Kim, Ung Sik Jin, Kyung Won Minn, and Hak Chang

Department of Plastic Surgery, College of Medicine, Seoul National University, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea

Correspondence should be addressed to Hak Chang; [email protected]

Received 11 September 2015; Accepted 5 November 2015

Academic Editor: Ahmad Jalili

Copyright © 2015 Byung Jun Kim et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background. Dermatofibrosarcoma protuberans (DFSP), a rare low-grade sarcoma of fibroblast origin, tends to extend in a finger- like fashion beyond macroscopic tumor margins. Therefore, incomplete removal and subsequent recurrence are common. This study aimed to determine the efficacy of wide local excision (WLE) for controlling local recurrence of DFSP. Methods. The medical records of 90 DFSP patients who received WLE at our hospital between June 1992 and January 2015 were retrospectively reviewed. WLE was conducted including a 3 cm (range, 1 to 5 cm) safety margin according to tumor size, location, and recurrence status. Clinical and tumor characteristics and surgical methods were evaluated for risk factor analysis and local recurrence-free survival. Results. DFSP occurred most often in patients in their 30s (30%) and on the trunk (51.1%). Five patients (5.5%) experienced local recurrence during the 43.4-month follow-up period. Recurrence was found at a mean of 10.8 months after WLE. Although no factors were significantly associated with recurrence, recurrences were more frequent in head and neck. Recurrence-free survival was 87% in 6 years and 77% in 7 years. Conclusions. WLE with adequate lateral and deep margins can effectively control local recurrence rate and is a simple and effective method to treat DFSP.

1. Introduction infiltrates the surrounding dermis and subcutaneous tissue as a pseudopod; therefore, incomplete removal is common Dermatofibrosarcoma protuberans (DFSP) is a rare locally owing to its irregular shape. Hence, the local recurrence aggressive mesenchymal tumor of fibroblast origin; it rate is high. Deciding the proper surgical margin for com- accounts for only 1% of all soft tissue sarcomas and <0.1% of plete resection is challenging. Therefore, many studies have all malignancies [1]. The incidence of DFSP is approximately shown the superiority of Mohs micrographic surgery (MMS) 3–5 cases per million persons [2–4]. It usually occurs in young compared to wide local excision (WLE) while considering adults (aged between 20 and 40 years), and there is no definite the local control of DFSP [10–13]. MMS has advantages predominance considering the sex of the patient [5]. in controlling the tumor burden microscopically, but it DFSP can occur in any part of the body, but the trunk is is laborious, technically demanding, expensive, and time- the most common area involved, followed by the extremities consuming, which are major disadvantages [14]. and the head and neck [6]. DFSP grows slowly, similar to Therefore, the purpose of this study was to determine the nodules that appear as hypertrophic scars or benign soft efficacy of WLE with a proper resection margin by using a tissue tumors without any definite symptoms. Delays in large amount of clinical data on DFSP at a single center in diagnosis are common because of the benign appearance and Korea. its rarity. DFSP rarely metastasizes to the regional lymph nodes or 2. Materials and Methods distant organs, with the possibility of metastasis being <5% [7, 8].However,DFSPshowslocallyaggressivebehaviorandthe The medical records of patients with DFSP confirmed on local recurrence rate is between 0% and 60% [9]. The tumor histologic analysis and who were treated at our hospital 2 BioMed Research International between June 1992 and January 2015 were retrospectively Head and Tumor ≥ Recurring reviewed under the approval of the institutional review neck? 5 cm? lesion? board (IRB number 1508-157-699). The patients who were diagnosed with DFSP but did not undergo WLE at our 5 cm hospital were excluded from this study. We examined the Yes medical records to obtain medical information of patients with DFSP,including the clinical and tumor characteristics as No 4 cm 4 cm well as the surgical methods and outcomes. This information Yes Yes piece was used to investigate the clinical course of DFSP and No the prognostic factors associated with recurrence after WLE. 3 No 3 3 cm No 3 cm cm cm Yes 2.1. Treatment Protocol. Most of the patients with DFSP Yes Yes were referred to our clinic from primary clinics, for definite No surgical treatment after incomplete excision or owing to 2 cm 2 cm No 2 cm recurrent lesions. Patients were categorized as primary versus recurrence cases. The patients were defined as having pri- Figure 1: The flow sheet shows how to determine resection margins. mary DFSP when histology confirmed DFSP and WLE was The area of involvement, tumor size, and recurrence status affect the performed at our hospital within 3 months after diagnosis, determination of the resection margin. irrespective of the place where diagnosis was confirmed. The patients were defined as having recurrence when the DFSP mass was surgically excised using excisional biopsy or WLE avoid depressive pigmented scars from skin grafts (Figure 3). methods, but new lesions were found and confirmed to be In case of positive resection margins on permanent biopsy, DFSP. reoperation was performed whenever possible to obtain a Preoperative diagnostic workups included a general lab- clear margin. Adjuvant treatment including radiation or oratory checkup, chest radiography, magnetic resonance chemotherapy was performed only for huge masses, recur- imaging (MRI) of the primary lesion, and positron emis- rent lesions, or inoperable cases. sion tomography (PET) to rule out the involvement of the Postoperative surveillance at the primary site and the regional lymph nodes or metastasis to distant organs. Bone regional lymph nodes was performed in 3 and 6 months by scintigraphy was also performed when bony tissue was likely using physical examinations, chest radiography, and ultra- to be involved, as observed on radiologic imaging. Biopsy sonography. MRI or PET was performed in highly suspicious examination slides that had been obtained at other hospitals casesofrecurrenceormetastasis.From1yearaftersurgery, were reconfirmed by the pathologists at our hospital. annual checkups were performed until 5 years by using the WLE was performed in all patients, with the patient under same methods. general anesthesia. The lateral margins were decided based onthesizeoftumors,thelocationofDFSP,recurrencestatus, 2.2. Statistical Analysis. The Kaplan-Meier method was used andthepatternofspreadingonMRIimaging.Thestandard to evaluate the status of recurrence. Univariate and multivari- resection margin was 3 cm. Another 1 cm was added for ate analyses were performed using the Cox regression test masses exceeding 5 cm (i.e., huge masses) or for recurrent to identify risk factors that were associated with recurrence. lesions. More conservative resection margins were employed Statistical analysis was performed using the SAS program forsmallerlesionsorforthoselocatedintheheadand (version 9.3; SAS Institute Inc., USA). 𝑃-values less than 0.05 neck in order to prevent aesthetic and functional impairment were considered statistically significant. (Figure 1). MRI helped us to determine the area of primary lesions. Subcutaneous extension of DFSP, which is hard 3. Results to be identified on physical examinations, can be easily discovered using MRI (Figure 2). Deep resection margins 3.1. Clinical Characteristics. Of 90 subjects included in the routinely included the muscle fascia except for primary present study, 53 (58.9%) were men and 37 (41.1%) were cases with smaller tumors. Intraoperative frozen biopsy was women. DFSP mostly occurred between the age of 20 and 50 usually conducted to identify the remnant tumor burden years, and the mean age at surgery was 34.6 years (range, 0– after excision with lateral and deep margins. If frozen biopsy 78 years). The age distribution showed a normal distribution revealed positive resection margins, additional WLE was pattern with a peak for patients in their 30s (Figure 4). performed from the point at which positive results were The mean duration from onset to surgery was 56.4 months. obtained. The trunk (51.1%) was the most common site involved, Immediate reconstruction was performed after WLE was followed by the upper extremities (20.0%), lower extremities completed,andfrozenbiopsyrevealednegativeresection (16.6%), and head and neck (12.2%) (Figure 5). Most tumors margins. Relatively small defects in abdomen can be repaired did not exceed 30 mm; 5 tumors were larger than 50 mm. by primary closure without difficulty. In larger defects that Seventy-seven (85.6%) patients visited our clinic with de novo do not permit primary closure, skin grafts or local flaps were tumors or for definitive treatment after incomplete excision employed. Free flaps can be another option for large defects from another hospital. Thirteen (14.4%) cases of recurrence or aesthetically important areas, such as the head and neck to were initially misdiagnosed as benign lesions; these patients BioMed Research International 3

▲

∗

▲

Figure 2: MRI image showing the subcutaneous extension of DFSP.These images help to determine the outline of DFSP.Arrow heads present primary lesion of DFSP in scalp. Asterisks present subcutaneous extension beyond the macroscopic tumor margin. MRI: magnetic resonance imaging, DFSP: dermatofibrosarcoma protuberans.

Figure 3: Intraoperative photos show WLE of DFSP and reconstruction with a local flap, skin graft, and free flap, respectively. WLE: wide local excision, DFSP: dermatofibrosarcoma protuberans. 4 BioMed Research International

0 0∼9 10∼19 20∼29 30∼39 40∼49 50∼59 60∼69 70∼79

Figure 4: Age distribution presents a peak in patients in their 30s with a normal distribution pattern.

3.3. Prognostic Factors. The mean follow-up period was 43.4 months (range, 0.2–282.4 months). Local recurrence was found in 5 patients (5.5%); the mean period between surgery and recurrence in these patients was 10.8 months (range, 3– 21 months). All recurrences occurred in the head and neck (2 patients), shoulder (1 patient), lower limb (1 patient), and inguinal area (1 patient); that is, none of them occurred on the trunk, which is usually the most common location. Additional WLE with skin grafts was performed in these patients, and adjuvant radiation therapy was performed in 2 patients. There were no cases of metastasis to the regional Head and neck Inguinal girdle lymph nodes or distant organs. The Cox regression test did Abdomen and chest Upper extremities not identify any significant risk factors among the clinical Back Lower extremities characteristics and surgical methods that were related to Shoulder girdle localrecurrenceonbothunivariateandmultivariateanalyses. Figure 5: Circular graph shows distribution of location where DFSP Kaplan-Meier survival analysis showed that the recurrence- developed. DFSP: dermatofibrosarcoma protuberans. free survival was 87% in 6 years and 77% in 7 years (Figure 6). underwent excision at another clinic without any histological 4. Discussion diagnosis (Table 1). In the present study, DFSP occurred most often in patients in their 30s (30%) and on the trunk (51.1%) without definite 3.2. Treatment Characteristics. All patients underwent WLE predominance in sex distribution. This corresponds well with with adequate margins. The mean lateral resection margin was 2.94 cm (range, 1–5 cm). In 58 patients (64.4%), the the observation of previous articles [5, 19, 21]. Our data tumor was resected including the deep fascia in order to revealed relatively low risk of local recurrence rate (5.5%) obtain adequate deep margins. Permanent biopsy revealed after WLE, which is compatible with those observed after positive results in 4 patients (4.4%). Additional WLE was MMS [24, 25]. performed in all of these patients to remove the tumor MMS is one of the most prevalent methods to treat skin cells completely. The reconstruction methods used were skin cancer or sarcoma. MMS has shown a low risk of local grafts in 38.9% of the patients, primary closure in 27.8%, recurrence as well as reduced positive resection margins. local flaps in 23.3%, and free flaps in 10.0%. There were no Previous studies have presented the superior outcomes of major complications after surgery, such as flap failure. Minor MMS to WLE regarding the local recurrence rate, although complications such as partial graft loss, local flap conges- no randomized controlled studies have been performed [10, tion, or wound dehiscence were treated with conservative 12, 13, 26, 27]. These studies showed recurrence rates of 0% management (Table 2). Six patients underwent postoperative to 6.6% after MMS, whereas the rates increased from 11.0% radiotherapy after tumor excision; 3 of them had huge tumors to 35% after WLE. Relatively high local recurrence rate after while the other 3 had a history of repeated recurrence. Two WLE is attributable to incomplete resection margin of DFSP. patients received chemoradiotherapy; 1 patient had a huge Akram et al. reviewed articles regarding recurrence rates tumor in the forehead that had recurred repeatedly despite after WLE performed during 2000–2012 [28]. The pooled WLE, and the other patient had DFSP in her upper lip recurrence rate was 8.5% in 1432 patients, and the lower that could not be excised with adequate margins. The mean recurrence rate was related to wider excision. A review article radiation dose was 57.0Gy (range, 50.0–63.0 Gy). by Pallure et al. showed that WLE with less than a 3 cm BioMed Research International 5

Table 1: Clinical characteristics of 90 patients with DFSP.

Factors Number of patients Percentage (%) Sex Male 53 58.9 Female 37 41.1 Male Female Total Age 0–9 3 0 3 3.3 10–19 4 6 10 11.1 20–29 15 6 21 23.3 30–39 15 12 27 30.0 40–49 9 6 15 16.7 50–59 6 6 12 13.3 60–69 1 0 1 1.1 70–79 0 1 1 1.1 Male Female Total Location Head/neck 5 6 11 12.2 Abdomen/chest 19 9 28 31.1 Back 9 9 18 20.0 Shoulder girdle 6 4 10 11.1 Inguinal girdle 1 1 2 2.2 Upper extremities 7 1 8 8.9 Lower extremities 6 7 13 14.4 Tumor size <30 mm 58 64.4 30–50 mm 19 21.1 >50 mm 55.6 Unknown 8 8.9 Clinical presentation Primary 77 85.6 Recurred 13 14.4 Number of recurrences 1 77.8 2 4 4.4 ≥3 22.2 Duration prior to surgery (onset) 1year 25 27.8 2years 9 10.0 3years 9 10.0 >3years 29 32.2 Unknown 18 20.0 DFSP: dermatofibrosarcoma protuberans. resection margin resulted in an increased recurrence rate the resection margins and the prognosis of DFSP (Table 3). [29]. We also removed the deep fascia in most of the cases (64.4%) We used 3 cm lateral resection margins but reduced except for the primary cases with smaller tumors. Many them to 2 cm or less for smaller lesions in the head and authors suggested the importance of deep margin control neck while considering aesthetics. In recurrence cases or in while describing clinical cases with deep tissue invasion [24, case of huge tumors, an additional 1 cm resection margin 30, 31]. Fields et al. [32] recommended removing the deep was obtained to prevent the risk of further recurrence. fascia to completely eliminate vertical infiltrating cells based Our protocol in determining the lateral resection margin is on the fact that tumor depth is associated with disease-free compatible with that used in previous studies that evaluated survival. According to Loghdey et al., achieving sufficient 6 BioMed Research International

Table 2: Surgical characteristics of 90 patients with DFSP. intraoperative frozen biopsy, and active additional treatment according to the result of biopsy accounts for lower recur- Factors Number of patients Percentage (%) rence rate. MRI is effective in determining the outline of Resection margin DFSP that grows beyond the macroscopic tumor margin [33– <30 mm 23 25.6 35]. MRI can give us visual clues regarding the tumor exten- 30–50 mm 51 56.7 sion area, tumor depth, and the relationship with adjacent ≥50 mm 6 6.7 tissues. In addition, intraoperative frozen biopsy is a useful Unknown 10 11.1 tool to minimize positive resection margin; it is also cost- Resection depth effective [36]. In our study, 4 cases (4.4%) showed positive Subcutaneous 13 14.4 margins on frozen biopsy; therefore, additional WLE was Deep fascia 58 64.4 performed. Permanent biopsy revealed negative results in all of these cases. Unknown 19 21.1 Other prognostic factors include old age, DFSP with a Surgical margin in biopsy high-grade fibrosarcomatous component (FS-DFSP), recur- Negative 85 94.4 rence, the involved site, increased mitosis, and positive Positive 4 4.4 microscopic margins [5, 26, 37, 38]. In our study, the head Unknown 1 1.1 and neck and the extremities were associated with high rates Reconstructive methods of local recurrence, although the result was not significant, Primary closure 25 27.8 possibly because of the relatively thin subcutaneous layer in Skin graft 35 38.9 these locations compared to the trunk. Relative conservative Local flap 21 23.3 treatment in the head and neck could be another cause of Free flap 9 10.0 recurrence. This corresponds with the observation made by Paradisi et al. [10]. Meticulous excision and close follow- DFSP: dermatofibrosarcoma protuberans. ups should be performed when WLE is performed in these areas. Well-designed MMS could be a better surgical option Survival function in treating DFSP of head and neck lesion, because MMS 1.0 allows greater preservation of normal healthy tissue and better cosmetic results can be expected [39]. Future studies are required to determine the relationship between the tumor 0.8 location and local recurrence. Our study failed to identify risk factors associated with local recurrence owing to the small number of patients with recurrence for statistical analysis. In 0.6 addition, we modified the resection margin according to the location, tumor size, and recurrence status; therefore, it was difficult to discover factors related to recurrence on univariate 0.4 analysis. Cum survival Cum We restrictively performed adjuvant therapy for huge tumors, of frequent recurrence, or when significant mor- 0.2 bidity is anticipated following WLE in functional and aesthetic aspects. The National Comprehensive Cancer Net- work guidelines also recommend limited use of adjuvant 0.0 radiotherapy or chemotherapy [9]. Protein tyrosine kinase inhibitors (e.g., imatinib mesylate) can be efficaciously used 0.00 50.00 100.00 150.00 200.00 250.00 300.00 fortreatingunresectableormetastaticDFSPinpatientswith 𝑡 Time (months) translocation between chromosomes 17 and 22 ( (17:22)). Imatinib mesylate was approved by the FDA for the treatment Survival function Censored of unresectable, metastatic DFSP in adults [40]. Although not used in our study, imatinib mesylate can be effectively used Figure 6: Survival curve represents the recurrence-free survival. for uncontrolled metastatic DFSP in patients with positive Median follow-up was 3.6 years. The 6-year recurrence-free survival cytogenetic study. and 7-year recurrence-free survival were 87% and 77%, respectively. This study presents the findings from a large number of cases of DFSP in a single center in Korea. We focused on the detailed surgical protocol and its prognosis in the long deep margins is important because of the nonconcentric term. WLE is a simple and effective method for treating extension pattern of DFSP and the limitation of standard DFSP, and local control can be improved comparable to vertical sections on histology [14]. MMS if adequate resection margins are established. However, In our study, 5 of 90 patients (5.5%) showed recurrence, this study had the limitation of being a retrospective study which is relatively lower than that observed in previous without a control group. Most of the patients were referred articles. Establishing proper resection margins using MRI, to our hospital by primary physicians after simple excision; BioMed Research International 7 Follow-up period (years) period (months) Recurrence 5.5 10.0 (5–14) 3.6 (0.1–23.5) (%) Overall recurrence rate NR 7.0 NR 8.4 (1.0–15.6) RT (6) Adjuvant treatment RT + C (2) 1.5 RT (34) 8.5 18.5 (3–129) 7.0 (0.6–23.8) 2.0 NR 0.0 NR 13.2 (2–28) > > 2.0 (2) 0.9–5.0 NR 27.0 31.8 9.6 2.0 (12), < < ≥ Resection margin (cm) Age (years) patients Number of (male/female) Table 3: Review of the literature regarding the clinical characteristics and surgical outcome in the treatment of DFSP. Chang et al. [18]Monnier et al. [4] 2004 2006 Retrospective Retrospective 60 (23/37) 66 (36/30) 36 (10–70) 32.8 (0–67) 3.1 (1.0–6.0) RT (3) 16.7 38.0 (1–100) 4.9 Popov et al. [19]Paradisi et al. [10]Heuvel et al.[20]Farma et al. [21]Cai et al. [22] 2007 2008 2010 2010 Prospective Retrospective 2012 Retrospective 38 Retrospective (16/22) 40 (22/18) 38 (19/11) Retrospective 204 (76/128) 44 (10–83) 45 38 (8–77) 41 (1–84) 223 2.0–5.0 2.0 (0.5–3.0) 2.0–3.0 38 3.1 (11–80) NR RT (9) RT (8) NR 13.0 1.0 10.5 48.5 (13–84) 0.0 NR NR (81, 11) 5.3 (0.1–17.5) 7.4 (1.0–21.8) 4.8 (2–10) NR 3.3 Hamid et al. [23]Goldberg et al. [11]Our study 2013 2015 Retrospective Retrospective 45 (30/15) 25 (8/17) 38.4 (7–65) 46.1 Retrospective (12–84) 2.0–3.0 90 (53/37) 2.8 34.6 (0–78) NR 3.0 (1.0–5.0) RT (1) 22.2 0.0 32.0 NR 5.7 (0.5–13) 9.0 AuthorsSmola et al. [15]Rutgers et al. [16] Publishing year 1991 Study 1992 design Retrospective Retrospective 20 11 NR NR 1.0–5.0 NR 30.0 NR 8.8 Stojadinovic et al. [17] 2001 Retrospective 14 (6/8) 37 (26–73) DFSP: dermatofibrosarcoma protuberans, RT: radiation, and NR:recorded. not 8 BioMed Research International

therefore, it was difficult to obtain accurate information about study of 34 cases,” AnnalesdeChirurgiePlastiqueetEsthetique, the characteristics of the primary tumor and the histologic vol. 50, no. 3, pp. 179–188, 2005. results. [13] L. Loss and N. C. Zeitouni, “Management of scalp dermatofi- In conclusion, the results of this study show the epi- brosarcoma protuberans,” Dermatologic Surgery, vol. 31, no. 11, demiology and clinical characteristics in Korean patients part1,pp.1428–1433,2005. with DFSP, which is compatible with those of previous [14] M. S. Loghdey, S. Varma, S. M. Rajpara, H. Al-Rawi, G. Perks, studies. WLE with adequate lateral and deep margins can be and W. Perkins, “Mohs micrographic surgery for dermatofi- effectively used to control local recurrence. brosarcoma protuberans (DFSP): a single-centre series of 76 patients treated by frozen-section Mohs micrographic surgery with a review of the literature,” Journal of Plastic, Reconstructive Conflict of Interests &AestheticSurgery,vol.67,no.10,pp.1315–1321,2014. [15]M.G.Smola,H.P.Soyer,andE.Scharnagl,“Surgicaltreatment The authors declare that there is no conflict of interests of dermatofibrosarcoma protuberans. A retrospective study of regarding the publication of this paper. 20 cases with review of literature,” European Journal of Surgical Oncology,vol.17,no.5,pp.447–453,1991. References [16] E. J. Rutgers, B. B. R. Kroon, C. E. Albus-Lutter, and E. Gortzak, “Dermatofibrosarcoma protuberans: treatment and prognosis,” [1] B. Llombart, C. Serra-Guillen,C.Monteagudo,J.A.L.Guerrero,´ European Journal of Surgical Oncology,vol.18,no.3,pp.241– and O. Sanmart´ın, “Dermatofibrosarcoma protuberans: a com- 248, 1992. prehensive review and update on diagnosis and management,” [17] A. Stojadinovic, A. Hoos, H. M. Karpoff et al., “Soft tissue Seminars in Diagnostic Pathology,vol.30,no.1,pp.13–28,2013. tumors of the abdominal wall: analysis of disease patterns and [2] V. D. Criscione and M. A. Weinstock, “Descriptive epidemiol- treatment,” Archives of Surgery,vol.136,no.1,pp.70–79,2001. ogy of dermatofibrosarcoma protuberans in the United States, [18] C. K. Chang, I. A. Jacobs, and G. I. Salti, “Outcomes of surgery 1973 to 2002,” JournaloftheAmericanAcademyofDermatology, for dermatofibrosarcoma protuberans,” European Journal of vol. 56, no. 6, pp. 968–973, 2007. Surgical Oncology,vol.30,no.3,pp.341–345,2004. [3]P.Rouhani,C.D.M.Fletcher,S.S.Devesa,andJ.R.Toro, [19] P. Popov, T. Bohling,¨ S. Asko-Seljavaara, and E. Tukiainen, “Cutaneous soft tissue sarcoma incidence patterns in the U.S.: “Microscopic margins and results of surgery for dermatofi- an analysis of 12,114 cases,” Cancer,vol.113,no.3,pp.616–627, brosarcoma protuberans,” Plastic and Reconstructive Surgery, 2008. vol. 119, no. 6, pp. 1779–1784, 2007. [4] D. Monnier, C. Vidal, L. Martin et al., “Dermatofibrosarcoma [20] S. T. Heuvel, A. Suurmeijer, E. Pras, R. J. Van Ginkel, and H. protuberans: a population-based cancer registry descriptive J. Hoekstra, “Dermatofibrosarcoma protuberans: recurrence is study of 66 consecutive cases diagnosed between 1982 and related to the adequacy of surgical margins,” European Journal 2002,” JournaloftheEuropeanAcademyofDermatologyand of Surgical Oncology,vol.36,no.1,pp.89–94,2010. Venereology,vol.20,no.10,pp.1237–1242,2006. [21] J. M. Farma, J. B. Ammori, J. S. Zager et al., “Dermatofibrosar- [5]C.J.McPeak,T.Cruz,andA.D.Nicastri,“Dermatofibrosar- coma protuberans: how wide should we resect?” Annals of coma protuberans: an analysis of 86 cases—five with metasta- Surgical Oncology,vol.17,no.8,pp.2112–2118,2010. sis,” Annals of Surgery,vol.166,no.5,pp.803–816,1967. [22]H.Cai,Y.Wang,J.Wu,andY.Shi,“Dermatofibrosarcoma [6] N. Angouridakis, P. Kafas, W. Jerjes et al., “Dermatofibrosar- protuberans: clinical diagnoses and treatment results of 260 coma protuberans with fibrosarcomatous transformation of the cases in China,” Journal of Surgical Oncology,vol.105,no.2,pp. head and neck,” Head&NeckOncology,vol.3,article5,7pages, 142–148, 2012. 2011. [23]R.Hamid,A.Hafeez,A.Darzi,I.Zaroo,H.Owais,andA. Akhter, “Dermatofibrosarcoma protuberans: role of wide local [7] O. Arican, S. Bakaris, E. Bulbuloglu, and F. Ezberci, “Myoid dif- excision,” South Asian Journal of Cancer,vol.2,no.4,pp.232– ferentiation and EMA expression in fibrosarcomatous dermato- 238, 2013. fibrosarcoma protuberans,” Acta Dermatovenerologica Alpina, Pannonica, et Adriatica,vol.15,no.1,pp.39–44,2006. [24]S.N.Snow,E.M.Gordon,P.O.Larson,M.M.Bagheri,M. L. Bentz, and D. B. Sable, “Dermatofibrosarcoma protuberans: [8] A. T. Vidimos, T. N. Helm, and F. A. Papay, Dermatofibrosar- a report on 29 patients treated by Mohs micrographic surgery coma Protuberans, Blackwell Scientific, Malden, Mass, USA, with long-term follow-up and review of the literature,” Cancer, 1998. vol. 101, no. 1, pp. 28–38, 2004. [9] S. J. Miller, M. Alam, J. S. Andersen et al., “Dermatofibrosar- [25] K. W. Dawes and C. W. Hanke, “Dermatofibrosarcoma protu- coma protuberans,” Journal of the National Comprehensive berans treated with Mohs micrographic surgery: cure rates and Cancer Network,vol.10,no.3,pp.312–318,2012. surgical margins,” Dermatologic Surgery,vol.22,no.6,pp.530– [10]A.Paradisi,D.Abeni,A.Ruscianietal.,“Dermatofibrosar- 534, 1996. coma protuberans: wide local excision vs. Mohs micrographic [26] H.M.GlosterJr.,K.R.Harris,andR.K.Roenigk,“Acomparison surgery,” Cancer Treatment Reviews, vol. 34, no. 8, pp. 728–736, between Mohs micrographic surgery and wide surgical excision 2008. for the treatment of dermatofibrosarcoma protuberans,” Journal [11] C. Goldberg, D. Hoang, M. McRae, C. Chung, D. J. Leffell, of the American Academy of Dermatology,vol.35,no.1,pp.82– and D. Narayan, “A strategy for the successful management of 87, 1996. dermatofibrosarcoma protuberans,” Annals of Plastic Surgery, [27]W.E.Love,S.A.Keiler,J.E.Tamburro,K.Honda,A.K. vol.74,no.1,pp.80–84,2015. Gosain, and J. S. Bordeaux, “Surgical management of congenital [12] R. Behbahani, P.Patenotre, N. Capon et al., “To a margin reduc- dermatofibrosarcoma protuberans,” Journal of the American tion in the dermatofibrosarcoma protuberans? Retrospective Academy of Dermatology,vol.61,no.6,pp.1014–1023,2009. BioMed Research International 9

[28] J. Akram, G. Wooler, and J. Lock-Andersen, “Dermatofibrosar- coma protuberans: clinical series, national Danish incidence data and suggested guidelines,” Journal of Plastic Surgery and Hand Surgery,vol.48,no.1,pp.67–73,2014. [29] V. Pallure, N. Dupin, and B. Guillot, “Surgical treatment of Darier-Ferrand dermatofibrosarcoma: a systematic review,” Dermatologic Surgery,vol.39,no.10,pp.1417–1433,2013. [30] M. Gattoni, R. Tiberio, L. Angeli et al., “Dermatofibrosarcoma protuberans: surgical treatment using the Tubingen technique (31 cases),” Annales de Dermatologie et de Ven´ er´ eologie´ ,vol.134, no. 1, pp. 31–34, 2007. [31] J. Wacker, B. Khan-Durani, and W.Hartschuh, “Modified mohs micrographic surgery in the therapy of dermatofibrosarcoma protuberans: analysis of 22 patients,” Annals of Surgical Oncol- ogy,vol.11,no.4,pp.438–444,2004. [32] R. C. Fields, M. Hameed, L.-X. Qin et al., “Dermatofibrosar- coma protuberans (DFSP): predictors of recurrence and the use of systemic therapy,” Annals of Surgical Oncology,vol.18,no.2, pp. 328–336, 2011. [33] W. C. Torreggiani, K. Al-Ismail, P. L. Munk, S. Nicolaou, J. X. O’Connell, and M. A. Knowling, “Dermatofibrosarcoma protuberans: MR imaging features,” American Journal of Roentgenol- ogy,vol.178,no.4,pp.989–993,2002. [34]G.G.Millare,N.Guha-Thakurta,E.M.Sturgis,A.K.El- Naggar, and J. M. Debnam, “Imaging findings of head and neck dermatofibrosarcoma protuberans,” American Journal of Neuroradiology,vol.35,no.2,pp.373–378,2014. [35] E. Ozmen, G. Guney,¨ and O. Algin, “Magnetic resonance imaging of vulvar dermatofibrosarcoma protuberans—report of acase,”Radiology and Oncology,vol.47,no.3,pp.244–246,2013. [36]R.U.Ashford,S.W.McCarthy,R.A.Scolyer,S.F.Bonar, R. Z. Karim, and P. D. Stalley, “Surgical biopsy with intraoperative frozen section. An accurate and cost-effective method for diagnosis of musculoskeletal sarcomas,” The Journal of Bone & Joint Surgery—British Volume,vol.88,no.9,pp.1207–1211, 2006. [37] W. B. Bowne, C. R. Antonescu, D. H. Y. Leung et al., “Der- matofibrosarcoma protuberans: a clinicopathologic analysis of patients treated and followed at a single institution,” Cancer,vol. 88,no.12,pp.2711–2720,2000. [38] M. Kim, C. H. Huh, K. H. Cho, and S. Cho, “A study on the prognostic value of clinical and surgical features of dermatofibrosarcoma protuberans in Korean patients,” Journal of the European Academy of Dermatology and Venereology,vol.26, no. 8, pp. 964–971, 2012. [39] C. Serra-Guillen,´ B. Llombart, E. Nagore et al., “Mohs micrographic surgery in dermatofibrosarcoma protuberans allows tumour clearance with smaller margins and greater preservation of healthy tissue compared with conventional surgery: a study of 74 primary cases,” The British Journal of Dermatology, vol. 172, no. 5, pp. 1303–1307, 2015. [40] G. A. McArthur, “Molecular targeting of dermatofibrosarcoma protuberans: a new approach to a surgical disease,” Journal of the National Comprehensive Cancer Network,vol.5,no.5,pp. 557–562, 2007. Hindawi Publishing Corporation BioMed Research International Volume 2015, Article ID 453091, 5 pages http://dx.doi.org/10.1155/2015/453091

Clinical Study Comparison of the Clinical Characteristics and Outcome of Benign and Malignant Eyelid Tumors: An Analysis of 4521 Eyelid Tumors in a Tertiary Medical Center

Yu-Yun Huang,1,2 Wen-Yih Liang,2,3 Chieh-Chih Tsai,1,2 Shu-Ching Kao,1,2 Wei-Kuang Yu,1,2 Hui-Chuan Kau,2,4 and Catherine Jui-Ling Liu1,2

1 Department of Ophthalmology, Taipei Veterans General Hospital, No. 201, Sec.2, Shih-Pai Road, Taipei 112, Taiwan 2School of Medicine, National Yang-Ming University, Taipei, Taiwan 3Department of Pathology and Laboratory Medicine, Taipei Veterans General Hospital, Taiwan 4Department of Ophthalmology, Koo Foundation Sun Yat-Sen Cancer Center, Taipei 112, Taiwan

Correspondence should be addressed to Chieh-Chih Tsai; [email protected]

Received 11 September 2015; Accepted 7 October 2015

Academic Editor: Emmanuella Guenova

Copyright © 2015 Yu-Yun Huang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

We retrospectively reviewed the clinical features and outcome of benign and malignant eyelid tumors from 1995 to 2015 in a tertiary medical center. Among 4,521 histologically confirmed eyelid tumors, 4,294 (95.0%) were benign tumors and 227 (5.0%) were malignant tumors. The mean age at diagnosis was significantly higher in patients with malignant lid tumors than those with benign lid tumors (72.5 and 55.4 years, resp., 𝑝 < 0.001). The most common benign eyelid tumors were intradermal nevus (21.1%), followed by seborrheic keratosis (12.6%) and xanthelasma (11.2%). The most common malignant eyelid tumors were basal cell carcinomas (57.8%), followed by sebaceous gland carcinomas (21.1%) and squamous cell carcinomas (10.1%). There was a relative male predominance (63.4% and 49.2%, resp., 𝑝 < 0.001) and higher recurrence rate (11.9% and 4.4%, resp., 𝑝 < 0.001)inmalignant lid tumors as compared with those of benign lid tumors. Twenty-two patients (9.7%) received orbital exenteration/enucleation. Eightpatients(3.5%)withmalignantlidtumorsdiedofdisease.Patientswitheyelidmelanomawereassociatedwithahighmortality rate (25.0%). It is important to differentiate between benign and malignant eyelid tumors, because they may cause cosmetic disfigurement and severe morbidity, especially in those with malignant eyelid tumors.

1. Introduction 2. Materials and Methods

Eyelid skin is not only the thinnest skin of the body but We retrospectively reviewed the medical records of all also among the most common sunlight-exposed areas of patients with histologically confirmed eyelid tumors who skins.Apartfromsubcutaneousfatlayer,eyelidcontains were treated at Taipei Veterans General Hospital between all other skin structures that can be the origin of vari- January 1995 and July 2015. Collected data included age, ous benign or malignant tumors. Eyelid tumors could be gender, location and size of the tumor, recurrence, and cosmetically disturbing to patients as well as diagnostically especiallythetreatmentandoutcomeofmalignanteyelid difficult for family physicians, dermatologists, and ophthal- tumors. The clinical characteristics and outcome difference mologists. Although there are some studies of eyelid tumors betweenbenignandmalignanteyelidtumorswerecompared. in the literature, most of them focused on the relative frequency of benign and malignant eyelid tumors [1–4]. This 2.1. Statistical Analysis. SPSS computer statistical software is the first study to investigate and compare the clinical (version 20.0; SPSS, Chicago, USA) was used for statistical features and outcome between benign and malignant eyelid analysis. Significant differences between two groups were tumors. studied using 2-tailed Fisher’s exact test and Mann-Whitney 2 BioMed Research International

Table 1: The demographic data and clinical features of benign eyelid tumors.

∗ ∗ Benign eyelid tumors Number (%) Gender (M/F) Mean age (years, range) Laterality (R/L/B) Location (U/L/B) (𝑛 = 4294) Intradermal nevus 905 (21.1%) 296/609 46.0 (8–95) 425/420/9 350/423/10 Seborrheic keratosis 540 (12.6%) 348/192 66.9 (21–93) 230/262/18 175/274/6 Xanthelasma 483 (11.2%) 144/339 52.4 (3–87) 199/217/46 245/172/8 Epidermal cyst 350 (8.2%) 237/113 59.6 (0.5–94) 150/167/9 200/74/4 Chronic inflammation 321 (7.5%) 169/152 55.3 (1–91) 134/150/9 126/128/1 Verruca vulgaris 255 (5.9%) 162/93 63.0 (8–93) 101/130/5 125/80/1 Squamous papilloma 147 (3.4%) 92/55 62.3 (6–100) 51/80/4 74/36/2 Fibrosis with granulation 115 (2.7%) 61/54 46.5 (1–90) 55/55/2 58/39/0 Skin tag 101 (2.4%) 63/38 63.1 (17–90) 44/42/3 52/34/0 Papilloma cutis 95 (2.2%) 51/44 60.3 (10–94) 50/44/1 53/32/4 Chalazion 80 (1.9%) 35/45 43.6 (1–93) 37/39/4 37/45/3 Benign keratosis 71 (1.7%) 40/31 62.0 (13–90) 33/37/1 37/25/0 Keratinous cyst 62 (1.4%) 31/31 60.1 (16–90) 23/32/3 33/17/2 Compound nevus 51 (1.2%) 20/31 27.6 (6–60) 23/24/0 20/27/1 Sebaceous gland hyperplasia 51 (1.2%) 27/24 58.3 (15–92) 20/27/1 24/21/0 Hemangioma 46 (1.1%) 22/24 48.2 (3–81) 21/24/0 24/19/0 Milia 38 (0.9%) 20/18 62.9 (17–91) 11/18/4 16/11/0 Lipoma 24 (0.6%) 13/11 62.2 (32–81) 8/7/4 7/10/0 Pilomatrixoma 23 (0.5%) 12/11 34.2 (2–82) 11/11/0 16/5/0 Bowen disease 5 (0.1%) 3/2 71.3 (61–84) 3/1/0 3/1/0 ∗∗ Others 531 (12.4%) 258/273 57.1 (1–95) 247/256/19 259/224/3 ∗ Totalnumbernotequalto100% of cases because of incomplete chart information or data missing. ∗∗ Others including neurofibroma, lentigo simplex, solar elastosis, calcinosis cutis, junctional nevus, trichilemmal cyst, trichoepithelioma, pyogenic granuloma, freckle, and molluscum contagiosum. test. A value of 𝑝 < 0.05 was considered statistically signifi- predominance among patients with malignant eyelid tumors cant. as compared with those of benign eyelid tumors (63.4% and 49.2%, resp., 𝑝 < 0.001). In addition, malignant lid tumors 3. Result tended to present as an ill-defined lesion (89.2% and 7.6%, resp., 𝑝 < 0.001) and locate in the lower eyelid (59.4% and Duringthe20-yearinterval,atotalof4,521eyelidlesions 42.9%, resp., 𝑝 < 0.001), as compared with those in benign from 4,243 patients with histopathologic confirmation were lid tumors. Most patients with malignant lid tumors required included in our study, with 4,294 (95.0%) benign tumors wide excision and reconstructive surgery. Fifteen patients and 227 (5.0%) malignant tumors. The demographic data (6.6%) underwent adjunct therapy including chemotherapy, andclinicalfeaturesofbenigneyelidtumorsareshownin radiotherapy, or combined chemoradiotherapy. Twenty-two Table 1. In the group of benign lesions, the most common patients (9.7%) received advanced orbital exenteration or diagnoses were intradermal nevus (21.1%), seborrheic kerato- enucleation. The tumor recurrence was significantly higher sis (12.6%), xanthelasma (11.2%), and epidermal cysts (8.2%). in patients with malignant eyelid tumors as compared with The demographic data, treatment, and outcome of malignant those with benign lid tumors (11.9% and 4.4%, resp., 𝑝< eyelidtumorsaresummarizedinTable2.Inthegroupof 0.001). Eight patients (3.5%) with malignant lid tumors died malignant tumors, the most common tumors were basal of disease. Among these, patients with eyelid melanoma were cell carcinoma (BCC, 57.8%), which were predominantly associated with a high mortality rate (25.0%). found on the lower eyelids (78.3%) and in male patients (68.7%). Sebaceous gland carcinoma (SGC) was the second 4. Discussion most common eyelid malignancy (21.1%) and showed a predilection for the upper eyelid involvement (59.6%) and The most important function of eyelids is designed to protect female predominance (58.3%). Squamous cell carcinoma theeyeball.Despitetheirsmallsurfacearea, they are among (SCC) was the third most common eyelid malignancy (10.1%) the most sunlight-exposed area of skins. The thin skin of the and showed a male predominance (69.6%). The other com- eyelids is particularly sensitive to various irritants and UV mon eyelid malignancy is melanoma (3.5%). Table 3 shows and is thus prone to develop eyelid tumors. Approximately the comparison of clinical characteristics between benign 5 to 10 percent of all skin cancers occur on the eyelids [5]. and malignant eyelid tumors. There was a relative male We demonstrated in this study that malignant eyelid tumors BioMed Research International 3 ma, invasive nonkeratinizing carcinoma, myeloid sarcoma, Adjunct treatment Recurrence Mortality 1/1/1 2/0 1 (10%) 1 (10%) 0/1/1 3/0 0 (0%) 1 (20%) 1/0/0 5/0 0 (0%) 2 (25.0%) 0/3/0 5/0 10 (20.8%) 1 (2.1%) 0/2/0 2/0 1 (50%) 0 (0%) 0/0/0 1/0 8 (6.1%) 0 (0%) 0/0/4 2/2 7 (30.4%) 3 (13.0%) Chemo/radiation/CCRT Exenteration/enucleation Number (%) Number (%) ∗ (U/L/B) Location ∗ (R/L/B) Laterality carcinoma, B-cell lymphoma, T-cell lymphoma, lymphoepithelial carcino (years, range) Mean age (M/F) Gender Table 2: The demographicdata, clinical features, treatment, outcome and of malignant eyelid tumors. 10 (4.4%) 8/2 71.7 (46–90) 5/4/0 5/1/0 Number (%) of cases because of incomplete chart information or data missing. % ) ∗∗ Others including small cell carcinoma, epidermoid carcinoma, merkel cell 𝑛 = 227 Totalnumbernotequalto100 Malignant eyelid tumors ( Basal cell carcinoma 131 (57.8%) 90/41 72.5 (21–93) 55/64/3 25/90/0 Sebaceous gland carcinomaSquamous cell carcinoma 48Melanoma (21.1%)Adenocarcinoma 23 (10.1%) 20/28Adenoid cystic carcinomaOthers 74.1 16/7 2 (0.9%) (44–91) 71.7 5 (2.2%) 0/2 23/25/0 8 (40–93) (3.5%) 4/1 53.0 27/19/1 8/15/0 6/2 (47–59) 73.8 67.4 7/6/0 (64–88) 1/1/0 (21–84) 1/4/0 2/6/0 1/0/0 0/1/0 3/2/0 angiosarcoma, and malignant proliferating pilar tumor. ∗ ∗∗ 4 BioMed Research International

Table 3: Comparison of demographic, clinical characteristics, and recurrence between benign eyelid tumors and malignant eyelid tumors.

Benign eyelid tumors Malignant eyelid tumors Type of tumor (𝑛 = 4294) (𝑛 = 227) 𝑝 value Number (%) Number (%) ∗ Gender Male 1976 (49.2) 144 (63.4) <0.001 Female 2040 (50.8) 83 (36.6) Mean age (years, range) 55.4 ± 20.0 (0.5–100) 72.5 ± 12.8 (21–93) <0.001 ∗∗ Location Upper only 2025 (55.9) 83 (40.1) < Lower only 1558 (42.9) 123 (59.4) 0.001 (U : L) Both 45 (1.2) 1 (0.5) ∗∗ Laterality Right only 1804 (44.7) 95 (43.8) Left only 1976 (48.9) 119 (54.8) 0.33 (R : L) Bilateral 258 (6.4) 3 (1.4) Recurrence 189 (4.4) 25 (11.0) <0.001 ∗ 278 patients of benign eyelid tumors had at least two histopathologic diagnoses because of 2 or more lesions in the eyelids, with 4294 lesions among 4016 patients of benign eyelid tumors. ∗∗ Totalnumbernotequalto100% of cases because of incomplete chart information or data missing. showed some distinctive demographic and clinical features our study. Tumor recurrence developed in about 11.9% of and treatment outcome as compared to benign lid tumors. cases of malignant eyelid tumors. Apart from infiltrated Most benign eyelid tumors in the present study occurred adenoid cystic carcinoma from lacrimal gland, most tumor in relatively young individuals as compared to malignant recurrence developed in patients with SCC (30.4%) and lid tumors, which was similar to other studies [2, 4]. Apart SGC (20.8%). Although SCC and SGC are less common from SGC and infiltrated adenoid cystic carcinoma, most than BCC, they may behave biologically more aggressive malignant eyelid tumors in current study showed a strong and are potentially lethal tumors. The mortality rate of male predominance (Table 2). However, there was no sig- malignant lid tumors in current study was 3.5%, and SCC and nificant sex distinction among patients with benign eyelid melanoma were the most common cause of death (Table 2). tumors. In addition, benign eyelid tumors occurred more Particularly, eyelid malignant melanoma remained the most commonly in the upper eyelid. Conversely, malignant lid high mortality rate (25%) in our studies. Treatment for tumors tended to locate in the lower eyelid, which could be cutaneousmelanomaoftenrequireswidesurgicalexcision attributed to most BCC with a predilection for the lower with histologic assurance of complete tumor removal. Till eyelid. In contrast to other malignant lid tumors, SGC often now, there is no generally accepted consensus regarding the originated in the upper eyelid (Table 2), reflecting the greater appropriate surgical margins for eyelid melanoma. Safety numbers of meibomian glands and glands of Zeis there. On margins for other cutaneous melanomas cannot be applied the other hand, SGC occurred more frequently in females. in eyelid melanoma for anatomic and functional reasons, In the literature, a higher incidence of SGC for females is becausesuchrecommendationsmayleadtosevereocular also suggested with a female to male ratio of 1.5 to 1.0 [6– complications, even loss of the globe and vision. Mouriaux 11]. The incidence of the malignant eyelid tumors is subject et al. suggest a “slow-Mohs” micrographic surgery for eyelid to geographical variation. Genetics and racial predisposition melanomas in order to preserve ocular integrity [17]. may also play a role [9]. BCC is the most common eyelid malignancy worldwide and accounts for 80–95% of all eyelid 5. Conclusion malignancy [5, 12]. However, in various Chinese studies, the percentage of BCC in all eyelid malignancy was not so Although benign lesions comprise the majority of eyelid much[4,9,13,14].Inourstudy,BCCremainedthemost tumors, it is important to differentiate between benign and frequently malignant eyelid tumors (57.8%). SGC is very rare malignant eyelid tumors by their demographic and clinical in Western literature. However, SGC was the second most characteristics. Early detection and appropriate treatment common malignant eyelid tumor in our series and accounted can reduce the risk of cosmetic disfigurement and morbidity for 21.1%, which was similar to some studies in Asia [9, 13–16]. induced by these tumors. In our series, malignant eyelid tumors often presented as an ill-defined lesion and usually required wide excision Conflict of Interests and reconstructive surgery that may cause cosmetic disfigurement and severe morbidity. Twenty-two patients (9.7%) None of the authors have any commercial interest in the required advanced orbital exenteration or enucleation in material mentioned herein. BioMed Research International 5

Acknowledgments [16]R.Sihota,K.Tandon,S.M.Betharia,andR.Arora,“Malignant eyelid tumors in an Indian population,” Archives of Ophthalmol- This study was partially supported by grants (104-2314-B-075 ogy,vol.114,no.1,pp.108–109,1996. -056 -MY2) from the Ministry of Science and Technology, [17] F. Mouriaux, A. Stefan, S. Coffin-Pichonnet, L. Verneuil, and P. Taiwan, and a grant (V104-C-092) from Taipei Veterans Rousselot, “What size of surgical margins for carcinoma of the General Hospital, Taiwan. eyelid?” Journal Franc¸ais d’Ophtalmologie,vol.38,no.2,pp.154– 158, 2015. References

[1] M. Deprez and S. Uffer, “Clinicopathological features of eyelid skin tumors. A retrospective study of 5504 cases and review of literature,” The American Journal of Dermatopathology,vol.31, no. 3, pp. 256–262, 2009. [2] G. Sotiropoulos, C. Gartzios, V. Raggos et al., “Eyelid tumors at the university eye clinic of Ioannina, Greece: a 30-year retrospective study,” Middle East African Journal of Ophthalmology, vol. 22, no. 2, pp. 230–232, 2015. [3] F. C. Gundogan, U. Yolcu, A. Tas et al., “Eyelid tumors: clinical data from an eye center in Ankara, Turkey,” Asian Pacific Journal of Cancer Prevention,vol.16,no.10,pp.4265–4269,2015. [4] M. Ho, D. T. L. Liu, K. K. L. Chong, H. K. Ng, and D. S. C. Lam, “Eyelid tumours and pseudotumours in Hong Kong: a ten-year experience,” Hong Kong Medical Journal,vol.19,no.2,pp.150– 155, 2013. [5] B. E. Cook Jr. and G. B. Bartley, “Epidemiologic characteristics and clinical course of patients with malignant eyelid tumors in an incidence cohort in Olmsted County, Minnesota,” Ophthal- mology,vol.106,no.4,pp.746–750,1999. [6]M.T.DoxanasandR.W.Green,“Sebaceousglandcarcinoma,” Archives of Ophthalmology,vol.102,no.2,pp.245–249,1984. [7]S.Kaliki,A.Ayyar,T.V.Dave,M.J.Ali,D.K.Mishra, and M. N. Naik, “Sebaceous gland carcinoma of the eyelid: clinicopathological features and outcome in Asian Indians,” Eye, vol.29,no.7,pp.958–963,2015. [8]M.M.K.Muqit,B.Foot,S.J.Walters,H.S.Mudhar,F.Roberts, and I. G. Rennie, “Observational prospective cohort study of patients with newly-diagnosed ocular sebaceous carcinoma,” British Journal of Ophthalmology,vol.97,no.1,pp.47–51,2013. [9] J.-K. Wang, S.-L. Liao, J.-R. Jou et al., “Malignant eyelid tumours in Taiwan,” Eye,vol.17,no.2,pp.216–220,2003. [10] C. Ni and P. K. Kuo, “Meibomian gland carcinoma. A clinicopathological study of 156 cases with long-period follow-up of 100 cases,” Japanese Journal of Ophthalmology,vol.23,no.4,pp. 388–401, 1979. [11] M. Zurcher,¨ C. R. Hintschich, A. Garner, C. Bunce, and J. R. O. Collin, “Sebaceous carcinoma of the eyelid: a clinicopathological study,” British Journal of Ophthalmology,vol.82,no.9,pp. 1049–1055, 1998. [12] C. W. Lober and N. A. Fenske, “Basal cell, squamous cell, and sebaceous gland carcinomas of the periorbital region,” Journal of the American Academy of Dermatology,vol.25,no.4,pp.685– 690, 1991. [13]S.T.Mak,A.C.-M.Wong,I.Y.-F.Io,andR.K.-K.Tse, “Malignant eyelid tumors in Hong Kong 1997–2009,” Japanese Journal of Ophthalmology,vol.55,no.6,pp.681–685,2011. [14] L. X. Xiao, B. Li, L. S. Xian et al., “Eyelid neoplasms in the Beijing Tongren Eye Centre between 1997 and 2006,” Ophthalmic Surgery Lasers and Imaging,vol.39,no.5,pp.367–372,2008. [15] S.-B. Lee, S.-M. Saw, K.-G. A. Hong, T.-K. Chan, and H.-P. Lee, “Incidence of eyelid cancers in Singapore from 1968 to 1995,” British Journal of Ophthalmology,vol.83,no.5,pp.595–597, 1999. Hindawi Publishing Corporation BioMed Research International Volume 2015, Article ID 629587, 10 pages http://dx.doi.org/10.1155/2015/629587

Research Article + Single-Fraction Radiotherapy for CD30 Lymphoproliferative Disorders

Michelle S. Gentile,1 Maria Estela Martinez-Escala,2 Tarita O. Thomas,3 Joan Guitart,2 Steven Rosen,4 Timothy Kuzel,4 and Bharat B. Mittal1

1 Department of Radiation Oncology, Northwestern University, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL 60611, USA 2Department of Dermatology, Northwestern University, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL 60611, USA 3Department of Radiation Oncology, Loyola University, Chicago, IL 60153, USA 4Division of Medical Oncology, Department of Medicine, Northwestern University, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL 60611, USA

Correspondence should be addressed to Bharat B. Mittal; [email protected]

Received 22 June 2015; Accepted 31 August 2015

Academic Editor: Emmanuella Guenova

Copyright © 2015 Michelle S. Gentile et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

+ Objectives.CD30 lymphoproliferative disorder is a rare variant of cutaneous T-cell lymphoma. Sustained complete response following first-line treatments is rare. This retrospective review evaluates the response of refractory or recurrent lesions to palliative radiation therapy. Methods. The records of 6 patients with 12 lesions, treated with radiation therapy, were reviewed. All patients + received previous first-line treatments. Patients with clinical and pathological evidence of symptomatic CD30 lymphoproliferative disorder, with no history of other cutaneous T-cell lymphoma variants, and with no prior radiation therapy to the index site were included. Results. The median age of patients was 50.5 years (range, 15–83 years). Median size of the treated lesions was 2.5 cm (range, 2–7 cm). Four sites were treated with a single fraction of 750–800 cGy (𝑛=3) and 8 sites were treated with 4000–4500 cGy in 200– 250 cGy fractions (𝑛=3). Radiation therapy was administered with electrons and bolus. Median follow-up was 113 months (range, 16–147 months). For all sites, there was 100% complete response with acute grade 1-2 dermatitis. Conclusions.Forrecurrentand + symptomatic radiation-na¨ıve CD30 lymphoproliferative disorder lesions, palliative radiation therapy shows excellent response. A single fraction of 750–800 cGy is as effective as a multifractionated course and more convenient.

1. Introduction histologyalonecanbeinsufficientandclinicalcourseisoften used to determine diagnosis and treatment [3, 4]. Primary cutaneous T-cell lymphomas (CTCLs) are relatively There are five histological subtypes of LYP withA rare, with an annual incidence of 7 in 1,000,000 [1]. Primary being the most common presentation and B, C, and D + + cutaneous CD30 lymphoproliferative disorders (LPD) rep- resembling MF, CALCL, and CD8 cytotoxic T-cell lym- resent 25–30% of CTCLs and are the second most common + phoma, respectively [4]. There is also a recently described form after mycosis fungoides (MF [2]). CD30 LPD can rare angioinvasive variant of LYP designated as histological + be divided further into lymphomatoid papulosis (LYP) and subtype E [5]. Immunohistochemistry often shows CD30 primary cutaneous anaplastic large-cell lymphoma (CALCL) expression with large pleomorphic or anaplastic T cells. LYP with substantial overlap between the two diagnoses resulting is a chronic indolent disease with recurrent papulonodular in a spectrum of disease. Although molecular markers and lesions that present over a course of years to decades and genetic rearrangements can be used to aid in diagnosis, may spontaneously regress after weeks to months. LYP has 2 BioMed Research International an excellent prognosis with a 5-year disease specific survival making assessment of response of LYP lesions challenging. of 100% [2]. Patients with LYP, however, are at greater risk In addition, there are no studies of single-fraction palliative of second cutaneous or nodal lymphoid malignancies that RT for LYP or CALCL. precede, follow, or are associated with other lymphomas such Since1999,wehavetreatedasmallseriesofpatients + as MF, cutaneous, or nodal anaplastic large-cell lymphoma with refractory or recurrent symptomatic CD30 LPD lesions andHodgkin’sLymphoma[3,6,7].Ithasbeendescribedthat using both multifractionated and single-fraction RT.This ret- LYP, Hodgkin’s Lymphoma, and CTCL can be derived from rospective analysis describes the largest series of patients with + a single T-cell clone and a t(8:9) genetic translocation may CD30 LPD treated with localized radiation for palliation. be involved in the pathogenesis of LYP or its progression to malignant disease [8]. + 2. Materials and Methods Similar to LYP, CD30 expressionisseenin>75% of CALCL cells, which have a large anaplastic, pleomorphic After approval by the Institutional Review Board, our depart- appearance [2]. CALCL presents as rapidly growing solitary ment records were examined and a comprehensive chart + or localized nodules that are rarely multifocal, with the review was performed yielding 6 patients with CD30 LPD appearance of large ulcerating tumors or thick plaques. who were treated with localized single or multifraction- Spontaneous complete resolution or partial regression is ated palliative RT to 12 individual lesions between October + commonly reported in >40% of patients [4]. Skin relapse 1999 and July 2012. Distinction among CD30 LPD and is common, with extracutaneous dissemination to mainly borderline cases can be challenging; thus patients were regional lymph nodes occurring in approximately 10% of carefully selected so that borderline diagnoses were excluded. patients. 10-year disease specific survival for patients without The European Organization for Research and Treatment of lymph node involvement is >90% [2]. Cancer (EORTC)/International Society for Cutaneous Lym- There have been many therapeutic approaches for LYP phoma (ISCL)/United States Cutaneous Lymphoma Con- including topical steroids, psoralen plus ultraviolet light sortium (USCLC) consensus recommendations on primary + therapy (PUVA), and low-dose methotrexate, which may cutaneous CD30 LPDwereusedtoconfirmthediagnosis show high response rates [9, 10]. CALCL lesions are often of all patients [4]. The TNM staging system for primary treated with radiation therapy (RT) or surgery for localized cutaneous lymphomas other than MF and Sezary´ syndrome disease or low-dose methotrexate. In the case of rapidly as proposed by the ISCL/EORTC was retroactively used for progressive or extracutaneous disease, treatment is with mul- staging [25]. Patient and tumor characteristics were assessed tiagent doxorubicin-based chemotherapy and more recently at initial consultation. All patients had disease refractory to brentuximab vedotin [2, 11]. There is often spontaneous prior topical and/or systemic treatment and no history of complete regression of smaller LYP lesions, but with larger other CTCLs or skin disorders and had not received prior lesions (>1-2 cm), a diagnosis of CALCL is more seriously RT to the index site. Date of last follow-up was defined considered and regression becomes less predictable. Relapse as the last encounter by a radiation oncologist, medical afterdosereductionorwithdrawaloftreatmentisatleast40% oncologist, or dermatologist where response to the treated and often much higher with LYP lesions in particular, and lesion had been documented. Patients were seen in follow- often these patients have lifelong disease with frequent relapse up 1 month following treatment and scheduled at 3–6-month [4]. Due to high relapse rates, maintenance therapy may be intervals thereafter. Death was confirmed by search of public used but may be accompanied by long-term complications death records. Pathology reports were reviewed with the including a higher incidence of nonmelanoma skin cancer dermatopathologist in order to determine the immunophe- and possible development of hepatic complications from notype and histological type for the six patients (see Figure 2). chronic methotrexate use [12]. In addition, misinterpretation Each lesion receiving RT was categorized based on its + of the clinical presentation of CD30 LPD for a more location. Parameters of RT assessed included total dose, aggressive disease (i.e., lymphoma, melanoma, or carcinoma) dose per fraction, energy, and bolus thickness. Response and the increased incidence of secondary lymphoid neo- was defined in a manner consistent with that put forth by plasms in LYP patients have led to treatment with systemic the EORTC/ISCL/USCLC consensus recommendations on + chemotherapy or even bone marrow transplantation [3, 13]. primary cutaneous CD30 LPD [4]; a CR was defined as 100% Kempf et al. [4] consensus guidelines for the treatment of clearance of the skin lesion treated, a partial response (PR) + CD30 LPD recommend consideration of RT for persistent, was defined as a reduction in lesion size of more than 50% but larger lesions greater than 2 cm. less than 100%, and stable disease (SD) was defined as a less Yu et al. [14] have described the treatment of CALCL than 50% reduction in size of the lesion. Relapse was defined with RT as an effective treatment modality. Outcomes of as any disease recurrence in those with CR. All patients had a LYP patients treated with RT, however, have been sparse CR; thus no patient or tumor characteristics were studied for and inconsistent. There are no known prospective studies, correlation with response. There were no identified relapses. and much of the published data is anecdotal or from small The RT regimen consisted of 750–800 cGy delivered in casestudies.Totalskinelectronbeamtherapy(TSEBT) a single fraction to 4 lesions or 200–250 cGy delivered in and localized fractionated RT have been used with variable multiple fractions for a total of 4000–4600 cGy to 8 lesions response [15–23]. There is limited information on RT details in the earlier years. En face electron technique was used andfollow-up[3,24].Manyofthesereportsincludedpatients for superficial lesions on flatter surfaces. Electron energy with synchronous or antecedent cutaneous lymphomas consisted of 10 or 12 MeV.Bolus was used for all of the lesions BioMed Research International 3

(a) (b)

+ Figure 1: (a) Patient with primary cutaneous CD30 lymphoproliferative disorder (LPD) of the left lower extremity. The gross lesion isa raised nodule with central ulceration and surrounding erythema. (b) The same patient at follow-up visit 8 months after completion of a single fraction of radiation therapy (RT) to 800 cGy. There is no clinical evidence of residual cutaneous lymphoma. All what remains is fibrotic tissue, which continues to fade. There was no evidence of recurrence at the last follow-up 27 months after treatment.

with a 0.5 or 1 cm thick material in order to increase radiation a multifractionated course of RT in the earlier years, the dosage to the skin. Electron dose was prescribed to the 90– median follow-up was 131 months (range, 66–146 months). 95% isodose line. All of the 6 patients were alive with disease at last follow-up with no evidence of relapse at the treatment site. 3. Results 4. Discussion Using the strict criteria described above, 6 patients with 12 + + localized, CD30 LPD lesions were treated with palliative RT. Cutaneous CD30 LPD is an indolent, recurrent variant Thisstudyconsistedof3femaleand3malepatientswith of CTCL that has been shown to be radiosensitive. Recent a total of 12 lesions. The median age was 50.5 years (range, consensus recommendations include surgical excision or RT 15–83 years) at initial time of RT treatment (Table 1). The for larger (defined as >2 cm in diameter) persistent lesions as median diameter of the lesion was 2.5 cm (range, 2–7 cm). an alternative approach to waiting for spontaneous regression + All patients had a history of biopsy proven CD30 LPD; [4]. In regard to this recommendation, however, there is little six of the 12 lesions had pathological confirmation while recent published evidence as to the clinical efficacy of local the remainder of patients were described as having lesions radiation or the recommended dose, fractionation scheme, that waxed and waned or recurrent papulonodular lesions technique, or long-term follow-up, especially with regard to refractory to first-line therapy consistent with their history LYP. Diagnosis by histology alone remains challenging and + of CD30 LPD. All lesions continued to progress following clinical presentation is often important. This small retrospec- first-line or other therapies and were symptomatic and none tive series represents the largest series to date specifically + of the lesions had evidence of spontaneous regression. One reporting localized RT outcomes for CD30 LPD using a patient received oral methotrexate prior to RT and another multifractionated and single-fraction approach. patient in the earlier years received CHOP chemotherapy A critical review of the literature (Table 2) showed that all prior to RT for a synchronous diagnosis of subcutaneous studies used a multifractionated course of RT for treatment Non-Hodgkin’s Lymphoma. All patients presented with gen- of LYP, with total doses of 8–40 Gy administered through eralized skin involvement consistent with T3 disease [25]. Of either TSEBT or localized superficial RT. Willemze et al. [15] the 6 lesions with pathological confirmation, all showed a treated one patient with two separate lesions. The first lesion + + CD4 /CD30 immunophenotype. Three lesions were of type was treated with TSEBR to a total dose of 40 Gy; the patient CLYPhistologyand1lesionwasoftypeALYPhistologywith experienced a CR but locally recurred within 3 months. A the remainder not specified. second lesion was treated with localized RT to a total dose of All patients had a CR to radiation (Table 2; Figure 1). RT 25 Gy; again the patient initially experienced a CR but locally was well tolerated, with the only recorded toxicity being grade recurred within 5 months. The patient went on to develop a 1-2 dermatitis. Median follow-up was 113 months (range, systemic lymphoma. Sanchez et al. [16] treated 4 of 31 patients 16–146 months) for the group as a whole. For the patients with LYP using various radiation therapies including TSEBR receiving a single fraction of RT, the median follow-up was to a total dose of 30 Gy and reported no response in 3 of 22.5 months (range, 16–37 months). For the patients receiving 4 patients. Kaufmann et al. [19] treated 1 of 2 patients with 4 BioMed Research International up (m) LR/follow- CR N/131 CR N/131 CR N/131 CR N/131 CR N/127 CR N/99 CR N/16 CR N/27 CR N/146 CR N/37 at RT site Response , , , − − − , − ,1cm ,1cm ,1cm ,1cm ,1cm ,1cm − − − − − − frac- 42/2, 45/2.5, 45/2.5, 45/2.5, 45/2.5, 45/2.5, 45/2.5, 95% IDL 90% IDL 90% IDL 90% IDL Total dose bolus/IDL 10 MeV e 1cmbolus, 1cmbolus, 1cmbolus, tion/energy/ 0.5 cm bolus, (Gy)/dose per 8/8, 12 MeV e 8/8, 12 MeV e 8/8, 10 MeV e bolus, 90% IDL bolus, 90% IDL bolus, 90% IDL bolus, 90% IDL bolus, 90% IDL bolus, 90% IDL 10 MeV e 10 MeV e 10 MeV e 10 MeV e 10 MeV e 10 MeV e 6c × IL IL Prior topical Oral/IL therapies acetonide IL steroids imiquimod methotrexate, methotrexate, triamcinolone + + + + + /CD30 /CD30 /CD30 /CD30 /CD30 + + + + + Immunophenotype type Histological diagnosis pathologic Clinical or of RT 4/7/03 Path. n/s CD4 Date of 3/01/06 Clin. 11/12/10 Path. n/s CD4 08/15/11 Path. n/s CD4 12/26/02 Clin. CHOP 12/26/02 Clin. 12/26/02 Path. C CD4 12/26/02 Clin. completion lymphoproliferative disorder (LPD) patients treated with radiation therapy (RT). + Left Left Left neck thigh axilla axilla groin Right Right Right Right Right lower elbow lateral region femoral Left calf 09/07/11 Path. C CD4 of lesion Location extremity of lesions treated with RT Number ∗ at T3b 6 TNM stage presentation Table 1: Characteristics of CD30 NHL lymphoma synchronous Synchronous subcutaneous Antecedent or at Sex/age (y) presentation Patient 1 M/48 n/a2M/60 T3b 1 3 F/83 Antecedent4 HL T3b F/53 2 n/a T3a 1 Upper lip 10/21/99 Path. A n/s BioMed Research International 5 up (m) LR/follow- CR N/18 CR N/66 at RT site Response ,1cm ,1cm − − frac- 40/2, 7. 5 / 7. 5 , Total dose bolus/IDL tion/energy/ (Gy)/dose per bolus, 95% IDL 12 MeV e 12 MeV e bolus, 90% IDL RT, Prior PUVA, Topical therapies acetonide prednisone, minocycline methotrexate, triamcinolone + /CD30 + Immunophenotype type Histological = electron, IDL = isodose line, CR = complete response, AWD = alive with disease, LR = local recurrence, − diagnosis pathologic Clinical or Table 1: Continued. of RT 7/23/12 Path. C CD4 Date of completion Left forearm of lesion Location of lesions treated with RT Number ∗ at TNM stage presentation lymphoma synchronous Antecedent or at Sex/age (y) presentation Staging per Kim et al., 2007 [25]. Patient 56 F/34 n/a M/15 n/a T3b 1 T3b 1 Left back 9/28/08 Clin. and N = none. ∗ HL = Hodgkin’s Lymphoma, NHL = Non-Hodgkin’s Lymphoma, n/s = not specified, ILintralesional, = e 6 BioMed Research International up (m) Follow- (m) Time to LR 100% Y 3 100% Y 5 0% N 100% N 14 100% Y n/s 100% N 100% N 4MeV RT details CR rate LR Localized RT 100% Y n/s 25 Gy, 100 kV Localized RT, TSEBR, 40 Gy, , − − − Other 30 Gy, PUVA topical TSEBR, 30/2 Gy, steroids, 90% IDL 6MeVe 6MeVe 6MeVe treatments 35/2–2.5 Gy, 30 mm HVL 0.5 cm bolus, 6/2 Gy, 15 kV, Localized RT 0% N Topical 5-FU, Localized RT, Localized RT, Localized RT, ∗ Topical mustard prednisone, prednisone, combination TNM stage at presentation chlorambucil, Mustine HCL, chemotherapy corticosteroids, Topical nitrogen procarbazine HCL vincristine, sulfate, photochemotherapy, n/s n/s T1bN0 Locations of treated lesion n/s T3bN0 Localized RT 100% N 36 Right forearm Left thigh T3bN0 Associated lymphoma Right finger HL n/s MF n/s T3aN0 MF n/s T3aN0 ML n/s Localized RT 0% N treated Number of LYP lesions Table 2: Studies including lymphomatoid papulosis (LYP) patients treated with radiation therapy (RT). RT Number of treated with LYP patients E11MFn/sT3bN0 R1 2LCL R4 R2 2 R1 2 Author [ref.] Study type Thomsen and Schmidt [21] Willemze et al. [15] Sanchez et al. [16] Sina and Burnett [22] Kaufmann et al. [19] BioMed Research International 7 70 up (m) 44.1 12, 48, 61, and Follow- (median) 3- 4.8; (m) year 20% DFS Time to LR Y6 12 80% N with 100% N 12 100% N 45 100% Y 0% (100% retreatment) , , ous T-cell lymphoma, PUVA = psoralen and − − ecurrence, and DFS = disease-free survival. scalp RT details CR rate LR Localized RT, Localized RT, Localized RT, supplemented 8/2 Gy, 100 kV 8/2 Gy, 100 kV Stanford, 36 Gy feet and 6/2 Gy, TSEBR/6 fields, 120 kV to apical TSEBR/modified 20/1 Gy, 120 kV to 0.5 cm bolus daily 36/1 Gy, 6 MeV e perineum, soles of 30/2 Gy, 6 MeV e Other topical CHOP PUVA, Topical received mustard steroids, nitrogen treatments concurrent n/s; 1 patient ∗ = electron, HVL = half value layer, LR = local r − T1bN0 T2aN0 TNM stage at presentation 3 × Table 2: Continued. n/s 0% Y n/s Right Lower abdomen Left flank T2aN0 forearm forearm, left Locations of treated lesion LCL Associated lymphoma CTCL/MF/ treated Number of LYP lesions RT Number of treated with LYP patients E 5 MF n/s R 3R66R4 R14 n/s LocalizedRT100%N n/s T1-3Nx R2 2 Retrospective staging according to Kim et al., 2007 [25]. Author [ref.] Study type Wilson et al. [18] J. Breneman and D. Breneman [17] Christensen et al. [24] Cabanillas et al. [3] Kaufmann et al. [20] Scarisbrick et al. [23] ∗ ultraviolet A therapy, n/a = not applicable, n/s = not specified, TSEBR total = skin electron beam radiotherapy, e E = editorial, R = retrospective, MF = mycosis fungoides, LCL = large-cell lymphoma, HL = Hodgkin’s Lymphoma, ML = malignant lymphoma NOS, CTCL = cutane 8 BioMed Research International

(a) (b)

Figure 2: (a) Dense dermal atypical lymphocytic infiltrate with mild epidermotropism. (b) Detail of the atypical lymphocytes arranged in sheets and mitotic figures are identified. Atypical lymphocytes are positive for (c) CD4 and (d) CD30.

localized RT to a total dose of 35 Gy in 2.0 Gy fractions using with 8 Gy in 2.0 Gy fractions; although there was a 50% 6 MeV electrons and reported a durable CR, although follow- CR rate, the first patient showed a CR with an additional up was not specified. Wilson et al. [18] reported that treatment 8 Gy. The remainder of studies lacked sufficient information of 3 of 161 patients with LYP/CTCL with TSEBR to a total on total dose, fractionation scheme, technique, or follow-up dose of 30 Gy resulted in a 3-year DFS of 20%; all patients [3, 21, 24]. In our study, there was a 100% CR rate, supporting + had relapsed by 4.8 months. J. Breneman and D. Breneman, our data for single-fraction RT for palliation of CD30 LPD. [17] in an editorial response to this study, shared anecdotal Moreover, Thomas et al. [26] have shown a 94.4% CR rate for results of 5 patients treated with TSEBR to a total dose of primarily MF lesions treated with a single fraction of localized 36GywithaCRrateof80%andnorelapseataminimum palliative RT at a mean follow-up of 41.3 months. + follow-up of 12 months. Kaufmann et al. [20] also reported a CD30 LPDismorelikelytobemultifocal,presentingas favorable outcome in one patient treated with localized RT to a recurrent, self-healing papulonodular eruption that often 30 Gy in 2.0 Gy fractions using 6 MeV electrons with a CR and spontaneously resolves without treatment over weeks to no recurrence at a follow-up of 45 months. Taken together, months. All patients were referred for treatment of lesions results for LYP treated with a multifractionated course of RT that had not shown spontaneous regression with continued resulted in a 69% CR rate, but relapse was approximately 45% growth following first-line or other therapies and, thus, were + at a follow-up of 3-4 months. concerning for a diagnosis of CD30 LPD. Given that all Inaddition,Yuetal.[14]showeda100%CRratefor8 patients had a durable CR at the RT site with sufficient follow- patients with CALCL treated with a multifractionated course up, these response rates are likely to be reflective of treatment of RT ranging from 34 to 44 Gy in 2.0 Gy fractions with a itself and not due to the spontaneous regression of the lesions. + medianfollow-upof12months.Otherretrospectivestudies A second criticism may be that, given the overlap of CD30 have shown excellent CR rates for CALCL patients; however, LPDwithCALCL,aswellasthetendencytowardprogression data for patients treated with RT alone is lacking, especially or concurrent MF or Hodgkin’s Lymphoma, unambiguous with regard to specific RT dose, technique, and long-term histological diagnosis may be difficult [5, 27–29]. follow-up [14]. There have been few studies where a few fractions of 5. Conclusion low-dose RT were given to treat LYP. Sina and Burnett + [22] treated 2 of 5 patients with a course of 6 Gy in 2.0 Gy CD30 LPD is a radiosensitive CTCL variant. In addition fractions,resultingin100%CRrateandnorelapseat14 to a multifractionated course of RT, a single fraction of and 36 months. Scarisbrick et al. [23] treated 2 of 4 patients 750–800 cGy is effective in inducing a durable CR with BioMed Research International 9 minimal acute side effects. Longer follow-up is necessary [3]F.Cabanillas,J.Armitage,W.C.Pugh,D.Weisenburger,and before conclusions regarding local control can be made M. Duvic, “Lymphomatoid papulosis: a T-cell dyscrasia with a especially for patients treated with a single fraction. This propensity to transform into malignant lymphoma,” Annals of study is the largest retrospective series reporting palliative RT Internal Medicine,vol.122,no.3,pp.210–217,1995. dose, technique, treatment outcomes, and long-term follow- [4]W.Kempf,K.Pfaltz,M.H.Vermeeretal.,“EORTC,ISCLand + up supporting palliative localized RT for symptomatic CD30 USCLC consensus recommendations for the treatment of pri- LPDrefractoryorrecurrenttoothertherapies. mary cutaneous CD30-positive lymphoproliferative disorders: lymphomatoid papulosis and primary cutaneous anaplastic Abbreviations large-cell lymphoma,” Blood,vol.118,no.15,pp.4024–4035, 2011. CALCL: Cutaneous anaplastic large-cell lymphoma [5]W.Kempf,D.V.Kazakov,L.Scharer¨ et al., “Angioinvasive lym- CR: Complete response phomatoid papulosis: a new variant simulating aggressive lym- DFS: Disease-free survival phomas,” American Journal of Surgical Pathology,vol.37,no.1, EORTC: European Organization for Research and pp. 1–13, 2013. Treatment of Cancer [6]M.W.Bekkenk,F.A.M.J.Geelen,P.C.vanVoorstVaderetal., + ISCL: International Society for Cutaneous “Primary and secondary cutaneous CD30 lymphoproliterative Lymphoma disorders: a report from the Dutch Cutaneous Lymphoma LPD: Lymphoproliferative disorders Grouponthelong-termfollow-updataof219patientsand LYP: Lymphomatoidpapulosis guidelines for diagnosis and treatment,” Blood,vol.95,no.12, pp. 3653–3661, 2000. MF: Mycosis fungoides [7] M. E. Kadin, “Current management of primary cutaneous CTCLs: Cutaneous T-cell lymphomas + PR: Partial response CD30 T-cell lymphoproliferative disorders,” Oncology,vol.23, no. 13, pp. 1158–1164, 2009. PUVA: Psoralen plus ultraviolet light therapy RT: Radiation therapy [8]T.H.Davis,C.C.Morton,R.Miller-Cassman,S.P.Balk,and SD: Stable disease M. E. Kadin, “Hodgkin’s disease, lymphomatoid papulosis, and cutaneous T-cell lymphoma derived from a common T-cell TSEBT: Total skin electron beam therapy clone,” The New England Journal of Medicine,vol.326,no.17, USCLC: United States Cutaneous Lymphoma pp. 1115–1122, 1992. Consortium. [9] G. Lange Wantzin and K. Thomsen, “PUVA-treatment in lymphomatoid papulosis,” British Journal of Dermatology,vol.107, Disclosure no. 6, pp. 687–690, 1982. This paper was presented in abstract form at the American [10] M. A. Paul, D. P. Krowchuk, M. G. Hitchcock, and J. L. Society of Clinical Oncology Annual Meeting 2012. Jorizzo, “Lymphomatoid papulosis: successful weekly pulse superpotent topical corticosteroid therapy in three pediatric Conflict of Interests patients,” Pediatric Dermatology,vol.13,no.6,pp.501–506,1996. [11] A. Broccoli, E. Derenzini, C. Pellegrini et al., “Complete res- The authors declare they have no conflict of interests. ponse of relapsed systemic and cutaneous anaplastic large cell lymphoma using brentuximab vedotin: 2 case reports,” Clinical Authors’ Contribution Lymphoma, Myeloma and Leukemia,vol.13,no.4,pp.493–495, 2013. Michelle S. Gentile carried out data collection and analysis [12] H. Fujita, T. Nagatani, M. Miyazawa et al., “Primary cutaneous and drafted the paper; Maria Estela Martinez-Escala partic- anaplastic large cell lymphoma successfully treated with low- ipated in data collection and analysis and helped to draft dose oral methotrexate,” European Journal of Dermatology,vol. thepaper;TaritaO.Thomascarriedoutdatacollectionand 18,no.3,pp.360–361,2008. analysis; Joan Guitart participated in project design, coordi- [13]S.Laube,F.Shah,andJ.Marsden,“Consequencesofmisdiag- nateddatacollection,andhelpededitthepaper;StevenRosen nosis of lymphomatoid papulosis,” European Journal of Cancer participated in project design and helped edit the paper; Care, vol. 15, no. 2, pp. 194–198, 2006. [14] J.B.Yu,J.M.McNiff,M.W.Lund,andL.D.Wilson,“Treatment Timothy Kuzel participated in project design and helped edit + the paper; Bharat B. Mittal conceived of the study, project of primary cutaneous CD30 anaplastic large cell lymphoma design, oversaw data collection and analysis, and helped edit with radiation therapy,” International Journal of Radiation Oncology Biology Physics,vol.70,no.5,pp.1542–1545,2008. the paper. All authors read and approved the final paper. [15] R. Willemze, C. J. L. M. Meyer, W. A. Van Vloten, and E. Schef- References fer, “The clinical and histological spectrum of lymphomatoid papulosis,” British Journal of Dermatology,vol.107,no.2,pp.131– [1] P. T. Bradford, S. S. Devesa, W. F. Anderson, and J. R. Toro, 144, 1982. “Cutaneous lymphoma incidence patterns in the United States: [16] N. P. Sanchez, M. R. Pittelkow, S. A. Muller, P. M. Banks, and a population-based study of 3884 cases,” Blood,vol.113,no.21, R. K. Winkelmann, “The clinicopathologic spectrum of lym- pp. 5064–5073, 2009. phomatoid papulosis: study of 31 cases,” Journal of the American [2]R.Willemze,E.S.Jaffe,G.Burgetal.,“WHO-EORTCclassi- Academy of Dermatology,vol.8,no.1,pp.81–94,1983. fication for cutaneous lymphomas,” Blood,vol.105,no.10,pp. [17] J. Breneman and D. Breneman, “Response to total skin irra- 3768–3785, 2005. diation for mycosis fungoides associated with lymphomatoid 10 BioMed Research International

papulosis: editorial,” International Journal of Radiation Oncol- ogy, Biology, Physics,vol.32,no.1,p.274,1995. [18]L.D.Wilson,D.L.Cooper,A.L.Goodrichetal.,“Impactof non-CTCL dermatologic diagnoses and adjuvant therapies on cutaneous T-cell lymphoma patients treated with total skin electron beam radiation therapy,” International Journal of Radia- tion Oncology, Biology, Physics,vol.28,no.4,pp.829–837,1994. [19] T. Kaufmann, L. Z. Nisce, and R. T. Silver, “Lymphomatoid papulosis:casereportofapatientmanagedwithradiation therapy and review of the literature,” American Journal of Clini- cal Oncology, vol. 15, no. 5, pp. 412–416, 1992. [20] T. P. Kaufmann, M. Coleman, and L. Z. Nisce, “Ki-1 skin lymphoproliferative disorders: management with radiation therapy,” Cancer Investigation,vol.15,no.2,pp.91–98,1997. [21] K. Thomsen and H. Schmidt, “Response to cutaneous T- cell lymphomas: the Sezary syndrome, mycosis fungoides and related disorders: editorial,” Archives of Dermatology,vol.113, no. 2, pp. 232–233, 1997. [22] B. Sina and J. W. Burnett, “Lymphomatoid papulosis: case reports and literature review,” Archives of Dermatology,vol.119, no. 3, pp. 189–197, 1983. [23] J. J. Scarisbrick, A. V. Evans, A. J. Woolford, M. M. Black, and R. Russell-Jones, “Regional lymphomatoid papulosis: a report of four cases,” British Journal of Dermatology,vol.141,no.6,pp. 1125–1128, 1999. [24] H. K. Christensen, K. Thomsen, and G. L. Vejlsgaard, “Lym- phomatoid papulosis: a follow-up study of 41 patients,” Seminars in Dermatology,vol.13,no.3,pp.197–201,1994. [25] Y.H. Kim, R. Willemze, N. Pimpinelli et al., “TNM classification system for primary cutaneous lymphomas other than mycosis fungoides and Sezary´ syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the Cuta- neous Lymphoma Task Force of the European Organization of Research and Treatment of Cancer (EORTC),” Blood,vol.110, no. 2, pp. 479–484, 2007. [26] T. O. Thomas, P. Agrawal, J. Guitart et al., “Outcome of patients treated with a single-fraction dose of palliative radiation for cutaneous T-cell lymphoma,” International Journal of Radiation Oncology Biology Physics,vol.85,no.3,pp.747–753,2013. [27]H.L.Liu,R.T.Hoppe,S.Kohler,J.D.Harvell,S.Reddy,andY. + H. Kim, “CD30 cutaneous lymphoproliferative disorders: the Stanford experience in lymphomatoid papulosis and primary cutaneous anaplastic large cell lymphoma,” Journal of the American Academy of Dermatology,vol.49,no.6,pp.1049–1058, 2003. [28] R. C. Beljaards and R. Willemze, “The prognosis of patients with lymphomatoid papulosis associated with malignant lymphomas,” British Journal of Dermatology,vol.126,no.6,pp.596– 602, 1992. [29] H. H. Wang, T. Myers, L. J. Lach, C.-C. Hsieh, and M. E. Kadin, “Increased risk of lymphoid and nonlymphoid malignancies in patients with lymphomatoid papulosis,” Cancer,vol.86,no.7, pp.1240–1245,1999.