Virchows Arch (2017) 471:467–489 DOI 10.1007/s00428-017-2176-1

ORIGINAL ARTICLE

Histiocytic and dendritic cell neoplasms: what have we learnt by studying 67 cases Fabio Facchetti 1 & Stefano Aldo Pileri 2,3 & Luisa Lorenzi 1 & Valentina Tabanelli 2 & Lisa Rimsza 4 & Stefania Pittaluga 5 & Stephan Dirnhofer 6 & Christiane Copie-Bergman 7 & Laurence de Leval 8 & Andreas Rosenwald 9 & Andrew Wotherspoon 10 & Falko Fend 11

Received: 7 April 2017 / Revised: 26 May 2017 / Accepted: 8 June 2017 / Published online: 10 July 2017 # Springer-Verlag GmbH Deutschland 2017

Abstract Tumors derived from histiocytic and dendritic cells encompass a large and heterogeneous group of neoplastic and reactive conditions, and their diagnosis is challenging both for pathologists and clinicians. Diagnosis is based on morphological and phenotypical findings, but hybrid features are not uncommon. Furthermore, recent studies uncovered the molecular mechanisms driving some of these tumors, improving diagnostic adequacy, and providing the basis for effective therapeutic breakthroughs. Sixty-seven cases were submitted to the accessory cell and histiocytic neoplasms session at the European Association of Haematopathology/Society for Hematopathology workshop 2016 held in Basel, Switzerland. The cases included histiocytic sarcomas (HS), Langerhans cell tumors (LCT), ErdheimChester disease, interdigitating dendritic cell sarcomas

(IDCS), indeterminate dendritic cell tumors (IND-DCT), follicular dendritic cell sarcomas, and blastic plasmacytoid dendritic cell neoplasms. Rosai-Dorfman disease and, more rare, conditions such as ALK-positive histiocytosis were also submitted. These cases illustrated classical and unexpected features at morphological, phenotypical, and molecular levels, providing a valuable compendium for pathologists confronting with these tumors. The paper summarizes the most notable features of every single group of diseases, with comments about the most challenging issues, in the attempt to provide practical indications for their recognition. Keywords Histiocytes . Dendritic cells . Tumors . Neoplasms

Fabio Facchetti and Stefano Aldo Pileri have equally contributed to this study. This article is part of the Topical Collection on Aggressive B-cell lymphomas and histiocytic neoplasias * Fabio Facchetti [email protected] * Stefano Aldo Pileri [email protected]

1

Section of Pathology, Department of Molecular and Translational Medicine, University of Brescia, Spedali Civili, 25123 Brescia, Italy

2

Unit of Hematopathology, European Institute of Oncology, 20141 Milan, Italy

3

Bologna University School of Medicine, Bologna, Italy

4

Department of Laboratory Medicine and Pathology, The Mayo Clinic, Scottsdale, AZ, USA

5

Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, Bethesda, MD, USA

6

Institute of Pathology and Genetics, University Hospital Basel, Basel, Switzerland

7

Department of Pathology, Hopital Henri Mondor, INSERM U955, Université Paris-Est, Creteil, France

8

Institute of Pathology, Lausanne University Hospital, Lausanne, Switzerland

9

Institute of Pathology, University of Würzburg and Comprehensive Cancer Center Mainfranken, Würzburg, Germany

10

Department of Histopathology, Royal Marsden Hospital, London and Sutton, UK

11

Institute of Pathology and Neuropathology and Comprehensive Cancer Center, University Hospital Tübingen, Tübingen, Germany

468

Introduction Neoplasms derived from histiocytes and dendritic cells are rare diseases [1, 2], but since their recognition is still challenging, their exact incidence remains to be determined. The normal cell counterparts of these tumors are different subsets of hematopoietic dendritic cells (e.g., Langerhans cells, interdigitating dendritic cells, plasmacytoid dendritic cells), mesenchymal-derived dendritic cells (e.g., follicular dendritic cells, fibroblastic reticulum cells), and histiocytes/macrophages. Notably, in the fourth edition of WHO classification and updates [1], blastic plasmacytoid dendritic cell neoplasm (BPDCN) was included among acute myeloid leukemia and related precursor neoplasms. This manuscript summarizes the results of the session dedicated to accessory cell and histiocytic neoplasms of the 2016 European Association for Haematopathology (EAHP) and Society of Hematology (SH) workshop, held in Basel. Sixtyseven cases of tumors derived from histiocytes and dendritic cells were submitted to the workshop, comprising histiocytic sarcoma (12 cases), Langerhans cell tumors (10 cases), indeterminate DC tumor (5 cases), and interdigitating DC sarcoma (3 cases), BPDCN (4 cases), and follicular dendritic cell sarcoma (18 cases); other cases were related to more rare diseases, and some of them were not true neoplasms (e.g., RosaiDorfman disease), but have shown significant overlapping features with neoplastic disorders. This paper summarizes the most notable features of every single group of diseases, with comments about issues which have been challenging for both submitters and panelists.

Histiocytic sarcoma Histiocytic sarcoma (HS) consists of elements provided with the morphologic and phenotypic characteristics of mature tissue histiocytes/macrophages [1, 3–7]. HS affects patients at any age, although commoner in adults, with a slight male predominance [1, 3–7]. It occasionally occurs in association with mediastinal germ cell tumors, malignant lymphoma, myelodysplasia, and leukemia [1, 3–7]. By definition, neoplastic proliferations associated with acute monoblastic leukemia are excluded from this setting [1, 3–6]. The tumor more often develops at extranodal sites (e.g., intestine, skin, soft tissue, and bone) and rarely shows a systemic presentation [1, 3–7], thus encompassing the term Bmalignant histiocytosis^ [1, 3–6]. Morphologically, HS consists of large cells with a diffuse, non-cohesive growth pattern [1, 3–7]. Intrasinusoidal diffusion may occur in the lymph nodes, liver, and spleen. Tumor cells are often pleomorphic, including giant cells or spindle cells; the cytoplasm is abundant and acidophilic; nuclear atypia ranges from mild to severe and mitoses are easily encountered in frankly atypical cases. A

Virchows Arch (2017) 471:467–489

variable number of reactive small lymphocytes, neutrophils, eosinophils, and plasma cells may be admixed. HS diagnosis is based on demonstration of one or more histiocytic marker expression, including CD68, CD163, CD4, CD11c, CD14, and lysozyme (Table 1) and negativity of B and T cell markers, CD30, EMA, CD21, CD23, CD35, CD1a, langerin/CD207, CD13, CD33, myeloperoxidase, cytokeratins, and HMB45. S100 protein is positive in half of the cases [1, 3–7]. At the ultrastructural level, tumor cells show numerous lysosomes, Birbeck granules, and cell junctions being absent [1, 6]. Clonal IG or TR rearrangements have been recorded in a few cases [8, 9] (see later), thus not excluding HS diagnosis. Due to limited experience, there is no treatment of choice [1, 4–6]. Systemic forms are aggressive with a few patients alive at 5 years despite the type of chemotherapy (CHOP, ICE, or ABVD) used [4, 5]. Tumors presenting as localized masses seem to have a better outcome following resection associated or not with adjuvant radiotherapy and/or chemotherapy [1, 4–6]. Response to vemurafenib was observed in one case showing BRAF (V600E) mutation [10]. Twelve cases of HS were submitted to the workshop, ten with classical features and two posing the problem of the differential diagnosis between HS and an exuberant reactive proliferation of histiocytes. Tumor cells were heterogeneous, medium-sized with rather monotonous ovoid nuclei (Fig. 1a) to large cells with marked nuclear pleomorphism (Fig. 1b). Multinucleated and xanthomatous cells were frequent (Fig. 1c). Inflammatory background was generally scarce and included lymphocytes and granulocytes; histiocytic markers (i.e., CD68, CD163) were evenly expressed in all cases (Fig. 2a). Notably, in two cases, cells with full-blown Langerhans cell phenotype could be detected, as confirmed by double immunostains (Fig. 2b). These cells formed distinctive aggregates in one case (case #149 by Dr. Alobeid) or were scattered throughout the HS in another (#307 by Dr. Gaulard); interestingly, similar features were also observed in one case labeled as Batypical histiocytic proliferation^ (#191 by Dr. Bayerl). It should be noted that a double histiocytic and Langerhans cell component was found also in cases classified as to Langerhans cell tumors (e.g., cases #177 by Dr. Swerdlow and #347 by Dr. Mcginnis), indicating that the distinction between HS and LC tumors can be challenging in some cases, and the final diagnosis should rely on the identification of the main cell component. This finding also implies that the biological boundary among histiocytic and dendritic cell tumors of myeloid derivation is not always sharply defined. Five cases of HS occurred in patients with a previous history of B cell neoplasm (Table 2) (two follicular lymphomas, #307, #372 by Dr. Chiu; one chronic lymphocytic leukemia #151 by Dr. Konoplev, one multiple myeloma #306 by Dr. Vesely, and one hairy cell leukemia #363 by Dr. Bruneau).

Virchows Arch (2017) 471:467–489 Table 1 Immunophenotypic markers most commonly used in the diagnosis of tumors derived from histiocytes/macrophages (H/M) and hematopoietic dendritic cells

469

Antigen

Main reactivity in normal cells

Diagnostic usefulness, pearls, and pitfalls

BRAF V600E

None

High sensitivity and specificity for cells carrying this mutation (clone VE1). The mutation itself is not specific for any disease and may occur also in epithelial cancers and melanoma

CD1a

LC, dermal DC (subset), IDC (subset) All DC and H/M

Required for diagnosis of LCH/LCS and IND-DCT

CD4

Positivity excludes IDCS Limited usefulness, due to wide expression Generally diffuse cytoplasmic stain, in contrast to T cells where it is membranous

CD14

H/M

CD68

H/M

H/M-derived tumors (less frequently than in monocytic leukemias) Most H/M-derived tumors show diffuse granular cytoplasmic reactivity; positivity in DC neoplasms more variable; BPDCN often negative

PDC

Clone PGM1 should be preferred to others since it does not stain myeloid cells CD123

PDC and activated H/M

CD163 CD207/langerin CD303/BDCA2

H/M LC, IDC (subset) PDC

Factor XIIIa HLA-DR

H/M H/M, DC

Lysozyme S100 protein

H/M LC, IDC, activated H/M

TCL1

PDC

High sensitivity and specificity for BPDCN Can be expressed in LCH and in H/M-derived tumors High sensitivity and specificity for H/M-derived tumors Defining LCH/LCS. Excludes IDCS High specificity for BPDCN H/M-derived tumors Paranuclear dot expression in LCH/LCS, IND-DCT, and IDCS. Useful to distinguish tumoral from reactive LC proliferations (see text) H/M-derived tumors Required for diagnosis of LCH/LCS, IND-DCT, IDCS Tumoral H/M can express S100 in variable number of cells Required for diagnosis of RDD High sensitivity and specificity for BPDCN

LC Langerhans cells, DC dendritic cells, IDC interdigitating dendritic cells, LCH Langerhans cell histiocytosis, LCS Langerhans cell sarcoma, IND-DCT indeterminate dendritic cell tumor, IDCS interdigitating dendritic cell sarcoma, H/M histiocytes/macrophages, PDC plasmacytoid dendritic cells, BPDCN blastic PDC neoplasm, RDD Rosai-Dorfman disease

Case #307 may be regarded as prototypic of this group: it occurred in a 76-year-old female with an 8-year history of follicular lymphoma (FL), who developed multiple lymphadenopathies and a skin lesion on the right thigh. A lymph node biopsy and a skin biopsy showed transformed FL and HS, respectively. The two lesions were molecularly related, carrying the same IGH rearrangement and BCL2/MYC double hit translocation, a previously unreported finding. Notably, in case #149 HS weakly expressed PAX5, an unexpected feature since the absence of such transcription factor is required for differentiation into myeloid or other non-B lineage cells [11]: possible transdifferentiation from an occult B cell lymphoma was hypothesized by the contributor. In all five cases, HS was diagnosed after the lymphoproliferative disease, with a time interval ranging from 1 to 26 years; it presented as a new single or multiple lesions and was unsuccessfully treated with lymphoid-oriented protocols

(velcade + dexamethasone #306, R-CHOP #307, rituximab + bendamustine #372), including BTK inhibitor in a patient with long-standing CLL (#151). In all cases, a clonal relationship between HS and the lymphoma could be proved (detailed in Table 2), that in the case following HCL (#363) was represented by common BRAF V600E mutation and various chromosomal alterations, although additional cytogenetic abnormalities were detected in the histiocytic tumor. Finally, case #315 (by Dr. Quinones) was unique, since the HS occurred in a patient with mediastinal germ cell tumor and chronic myelomonocytic leukemia; the existence of a common precursor was suggested by the demonstration of TP53 mutation in all three neoplasms and identical chromosomal aberrations in HS and the mediastinal germ cell tumor. Druggable molecular alterations affecting BRAF and MAP2K1 were reported in four cases of HS (Table 3). In addition to case #363, BRAF V600E mutation was present

470 Fig. 1 Histiocytic sarcoma. Cytology may be monomorphic as in case #149 (a, by Dr. Alobeid) or can display nuclear pleomorphism as in case #223 (b, by Dr. Ganapathi), along with xanthomatous histiocytes (c)

Fig. 2 Histiocytic sarcoma. Diffuse expression of the histiocytic marker CD163 (a, case #215, Dr. Quinones). Langerhans cell component was admixed to the HS proliferation in case #307 (b: double immune stain for CD163 in blue and CD207 in brown, by Dr. Gaulard). Strong PDL1 expression in case #234 is shown in c (by Dr. Zhao)

Virchows Arch (2017) 471:467–489

Virchows Arch (2017) 471:467–489 Table 2 Cases of histiocytic/ dendritic cell tumors associated with lymphomas

471

Case

Diagnosis

Site

Time interval

Molecular features Similar

151

1. CLL

1. PB

6 years

306

2. HS 1. PCM

2. BM, LN, spleen 1. BM

1 year

IGH, IGK clonality

2. HS

2. Submandibular mass 8 years

t(11;14) IGH-CCND1 fusion IGH clonality

307

1. FL

1. LN

2. HS

2. Skin

372

1. FL

1. LN

1 year

363

2. HS 1. HCL

2. LN 1. PB

26 years

2. HS

2. LN

177

1. CLL

1. PB 2. Breast mass

268

2. LCH + Histiocytic proliferation with osteoclast-like cells 1. cHL

282

2. LCH 1. FL

362

2. LCH 1. LCH 2. Accelerated CLL+ LCH

2. LN

Different

IGH clonality

BCL2 and MYC rearrangements BCL2 rearrangement BRAF V600E Some cytogenetic alterations

Months

IGH, IGK clonality

LN

0

n.a.

1. LN

5 years

2. Skin 1. LN

t(14;18) IGH-BCL2 fusion

4 years

IGH clonality

Additional cytogenetic alterations in HS [88]

BRAF V600E in LCH

CDKN2A/B deletion Mutations of NRAS, MAP2K1, NOTCH1, KMT2D t(14;18) IGH-BCL2 fusion

122

1. FL

1. LN-BM-skin-breast

10 years

269

2. LCS 1. T-ALL

2. LN 1. PB

2 years

2. LCS

2. Tonsil

TRG and TRB clonality

339

1. NHB

Spleen

6 months

Deletions: TLX3 at 5q35, TRA/TRD at 14q11, TRG at 7p14 n.a.

345

2. IND-DCT 1. IND-DCT

LN

0

n.a.

274

2. SBC-LPD 1. IDCS

LN

0

IGH, IGK clonality

IGH clonality in FL

BRAF V600E in IND-DCT

2. CLL BM bone marrow, CLL chronic lymphocytic leukemia/small lymphocytic lymphoma, cHL classical Hodgkin lymphoma, FL follicular lymphoma, HCL hairy cell leukemia, HS histiocytic sarcoma, IDCS interdigitating dendritic cell sarcoma, IND-DCT indeterminate dendritic cell tumor, LCH Langerhans cell histiocytosis, LCS Langerhans cell sarcoma, LN lymph node, n.a. not available, NHB non-Hodgkin B cell lymphoma, not otherwise specified, PB peripheral blood, PCM plasma cell myeloma, SBC-LPD small B cell lymphoproliferative disease, TALL T cell acute lymphocytic leukemia

472 Table 3 Molecular abnormalities reported in the workshop cases

Virchows Arch (2017) 471:467–489

Case

Diagnosis

Somatic mutations tested

223 243 315

HS HS HS

BRAF V600Ea BRAF V600Ea, b TP53c

363

HS

BRAF V600Eb [88]

372

HS

BRAF V600Ea

HS

BCL2, BCL10, CDKN1B, KIT, MAP2K1c BRAF V600Ea, d

377 149

HS + LCS

a

BRAF V600E

Other genetic alterations

Chr1: trisomy with LOH 1p36.33-p22.3; Chr2: LOH 2q; Chr5: LOH 5q13.2-q35.3; Chr13: loss of 13q; Chr17: LOH 17p13.3-p13.1; Chr21: trisomy

CLIP2-BRAF fusion [45] Complex karyotypic abnormalities: 48, XX, +7, +12, der (14) t (11;14) (q11;p11.1), add (14) (q32), +18, der (22) t (1; 22) (q21; q13) [15]/46, XX [5] 3 copies of CCND1 (27.1%) and ETV6 (23.1%) consistent with trisomy 11q and 12 Gains: 1q23.1q42.3, 7p22.3q36.3, 11q13.2q25, 12p13.33q24.33

177 268

LCH LCH

BRAF V600Ea, b BRAF V600Ea, c

Losses: 22q13.2q13.33, Xp22.33q28 13q deletion

BRAF V600Eb 282 347

LCH LCH

MAP2K1c BRAF V600Ea BRAF V600Ea, c

370 139 269

LCH LCS LCS

BRAF V600Eb BRAF V600Ea, b BRAF V600Eb, c BRAF V600Ea, b

368

LCS

BRAF V600Eb

156 200 345

IND-DCT IND-DCT IND-DCT

BRAF V600Kb BRAF V600Ea BRAF V600Ea, b, c BRAF V600Ea, c b

BRAF V600E

Deletions of TLX3 at 5q35, TCRA/TCRD at 14q11, and TCRG at 7p14

Negative for del (5q), del (7q) or monosomy 7, del (17p), del (20q) and trisomy 8

RUNX1c 159 274 216

299 204

IDCS IDCS DCS, NOS BF DCS, NOS BPDCN

TP53c, e BRAF V600Ea BRAF V600Ea BRAF V600Ea, b ATM S1691Rb, f PBRM1 CICg

371 259 316 387

ALK+ H FDCS FDCS FDCS

BRAF V600Ec, h BRAF V600Ec

ALK translocation (break apart probe)

257

ECD

BRAF V600Ea, b

Karyotype: 41, XX, −5, −6, −9, −14, add (15) (p10), −16, add (17) (p11.2), −22, −22+r, +mar [9]/46, XX [9]

Virchows Arch (2017) 471:467–489

473

Table 3 (continued) Case

Diagnosis

Somatic mutations tested

Other genetic alterations

267 183

ECD RH

BRAF V600Eb BRAF V600Eb

KIF5B-ALK fusioni [45] IGH rearrangement

121

RDD

KRASb BRAF V600Eb

188

RDD

KRAS K117Nc [49] BRAF V600Eb

340

RDD

BRAF V600Ea

Mutated genes are uderlined. HS histiocytic sarcoma, LCH Langerhans cell histiocytosis, LCS Langerhans cell sarcoma, IND-DCT indeterminate dendritic cell tumor, IDCS interdigitating dendritic cell sarcoma, DCS, NOS dendritic cell sarcoma, not otherwise specified, BF blastoid features, BPDCN blastic plasmacytoid dendritic cell neoplasm, ALK+ H ALK-positive histiocytosis, FDCS follicular dendritic cell sarcoma, ECD Erdheim-Chester disease, RH reticulohistiocytosis, RDD Rosai-Dorfman disease a

Evaluated by immunohistochemistry

b

Evaluated by direct sequencing

c

Evaluated by pyrosequencing or next generation sequencing

d

DNA sequencing library of 405 genes

e

In-house NGS myeloid panel including 54 genes

f

NGS panel including 46 genes

g

Targeted sequencing panel including 405 cancer-related genes

h

NGS panel including 46 genes (ABL1, AKT1, ALK, APC, ATM, BRAF, CDH1, CSF1R, CTNNB1, EGFR, ERBB2, ERBB4, FBXW7, FGFR1, FGFR2, FGFR3, FLT3, GNA11, GNAQ, GNAS, HNF1A, HRAS, IDH1, JAK2, JAK3, KDR, KIT, KRAS, MET, MLH1, MPL, NOTCH1, NPM1, NRAS, PDGFRA, PIK3CA, PTEN, PTPN11, RB1, RET, SMAD4, SMARCB1, SMO, SRC, STK11, TP53, VHL)

i

Evaluated by RNA sequencing

also in case #243, CLIP2-BRAF gene fusion was detected in case #377 (by Dr. Ozkaya), and a targetable MAP2K1 mutation was reported in case #372. Interestingly, a successful Btargeted^ treatment was obtained in two of these cases (#243 and #372). In terms of potential therapeutic targets and in line with previous reports [12], the expression of the programmed death ligand 1 (PD-L1) was explored in all HS, either by the submitters or by the panelists, and it was detected strong and diffuse in three of them (#243, #315 and #223 by Dr. Ganapathi). In analogy with other solid tumors (e.g., lung cancer), the immunohistochemical demonstration of PD-L1 (Fig. 2c) might provide therapeutic options for these patients [13, 14]. Two cases submitted to the workshop as HS (case #191 and #304 by Dr. Czadar) were respectively regarded by the panelists as Batypical histiocytic proliferation^ and Breactive histiocytic population.^ The first case was particularly challenging, and it occurred in the calcaneus of a 12-year-old boy, histiocytic cells displayed pleomorphism and were intermingled with numerous granulocytes, and the lesion spontaneously regressed. The second case occurred in the same location (small bowel) of a previously treated double hit diffuse large B cell lymphoma and was composed of bland histiocytes with no mitoses and proliferative activity, as demonstrated by

double immunostain for CD163 and Ki-67. Moreover, the panel was unable to demonstrate any translocation in these histiocytes. The former of these cases highlights the difficulties in predicting the clinical outcome based on pure morphology. According to the WHO classification [1], the diagnosis of each entity requires the amalgamation of clinical data, morphology, phenotype, and molecular characteristics. This along with the statement of Emile et al. that Bmalignant histiocytosis diagnosis should only be confirmed in patients with rapidly progressing tumours^ [2], prompted us to exclude HS in this case. Nevertheless, one should remind that patients with HS but showing clinically localized disease and small primary tumors may have a more favorable long-term outcome [1, 6].

Langerhans cell tumors Neoplastic proliferations of Langerhans cells (LC) include LC histiocytosis (LCH) and LC sarcoma (LCS). In both entities, tumor cells are clonal, express CD1a, CD207, and S100 protein, and show Birbeck granules by ultrastructural examination. These features strongly support their origin form epidermal LC [1], although the possibility that LCH derives from different cell sources, including a CD1c+/CD207+ myeloid progenitor, has been considered [15].

474

Tumor cell proliferation in LCH is driven by extracellular signal-regulated kinase (ERK) activation [16], which in about 50% of cases is triggered by BRAF V600E mutation [17] or by activating mutations of MAP2K1, MAP3K1, or ARAF in a fraction of BRAF wild type cases [18, 19]. Recent data support a relationship between stages of maturation of the cell in which the mutational hit occurs and extent of the disease, explaining the broad spectrum of clinical manifestations [15, 20]. In this model, oncogenic signals affecting bone marrow stem cells or committed precursors would induce an aggressive, disseminated disease, while low-risk single lesions would be caused by somatic mutations occurring in mature LC [20]. Independently from disease spread, in LCH, tumor cells show similar features with grooved folded or lobulated nuclei, fine chromatin, inconspicuous nucleoli, and thin nuclear membranes. In LCS, tumor cells are significantly atypical and morphologically hardly recognizable as Langerhans cells. Moreover, according to Emile et al., the diagnosis of LCS should be better established on combination of histological (Bmajor nuclear atypia^) and clinical criteria (Brapid tumor progression^) [2]. Eleven cases of LC tumors (6 LCH and 5 LCS) were submitted to the workshop and illustrate the main diagnostic issues of this group of diseases. Case #370 (by Dr. Kuzu) represented a nodal LCH with an exclusive intrasinusoidal growth mimicking reactive sinus histiocytosis [21]; it occurred in a young woman with deep and superficial lymphadenopathy and multiple lytic bone lesions. The intrasinusoidal ovoid cells showed morphological features of LC and were intermingled with eosinophils (Fig. 3a, b); they displayed a complete LC phenotype (S100+, CD1a+, CD207+) with negativity for histiocytic markers (CD68 and CD163) (Fig. 3c). In four patients, LCH occurred in association with previous or contemporary diagnosis of lymphoproliferative disease, namely, nodular sclerosis Hodgkin lymphoma (HL) (#268 by Dr. Zhang), chronic lymphocytic leukemia (CLL) (#177, #362 by Dr. Xerri), and follicular lymphoma (FL) (#282 by Dr. Ko) (Table 2). In case #268, LCH formed aggregates at the periphery of or intermingled with HL nodules (Fig. 3d), in case #362 LCH diffusely replacing a lymph node was diagnosed 4 years before CLL diagnosis and persisted as small aggregates during lymphoma progression into accelerated CLL; finally, in case #282, LCH occurred as diffuse dermal infiltrate 5 years after nodal FL diagnosis. Significance of the association between LCH and lymphomas is discussed in a following paragraph. In all submitted cases, LCH diagnosis was based on coexpression of CD1a, S100, and CD207, with the latter two markers being more uneven compared to CD1a. A partial phenotypic shift towards a histiocytic phenotype may occur especially in long-standing lesions, likely due to the influence

Virchows Arch (2017) 471:467–489

of the associated inflammatory microenvironment [15, 22]. In two submitted LCH cases (#177 and #347), the diagnosis was remarkably challenging, due to extensive histiocytic differentiation. In case #347, nodal LCH was associated with small histiocytic nodules occurring in the spleen, the latter showing marked erythrophagocytosis, positivity for S100, CD68, and CD163, but not CD1a and CD207; notably, this Bhistiocytic^ component shared the same BRAF V600E mutation with LCH. In case #177 (breast mass in a 57-year-old man recently diagnosed with CLL), LCH was associated with a distinct and dominant histiocytic proliferation containing numerous osteoclast-like giant cells (Fig. 3e). A clonal relationship between the two was supported by the presence of the same IGH and IGK rearrangements, but BRAF V600E mutation was detectable only in LCH (Fig. 3f). LCS cases showed diffuse involvement of the lymph node (#122 by Dr. Gao), tonsil (#269 by Dr. Wright), and skin (#139 by Dr. Hock and #368 by Dr. Feichtinger) by large atypical cells, with abundant eosinophilic cytoplasm and high proliferation rate. A progression from LCH was suggested by case #139, where cutaneous LCS occurred 1 year after LCH diagnosis in skin and lymph node, and by case #368, where LCH and LCS coexisted in the same tissue (Fig. 4a, b). CD1a and S100 were generally expressed in the majority of tumor cells in all cases, while CD207 was highly variable (Fig. 4c, d) and, in case #139, it was completely lacking, despite Birbeck granules were identified by the authors. Since Birbeck granule formation in Langerhans cells depends on CD207 [23], lack of immunoreactivity in this case could depend from gene mutation resulting in structure changes of the protein [24]. This finding addresses the issue regarding the differential of LCS from indeterminate dendritic cell tumor (IND-DCT) on pure phenotypical bases. Two cases of LCS had a history of lymphoma, respectively, follicular B cell lymphoma (case #122) and T cell lymphoblastic leukemia/lymphoma (#269); tissues involved by LCS did not show residual lymphoid neoplasm, but in both cases, there were molecular features of clonal relationship (see later); interestingly, in #269, LCS expressed CD34 (but not TdT and other T cell markers), as also proved by double CD1a-CD34 stain performed by the panel (Fig. 4e, f).

Indeterminate DC tumor/histiocytosis Indeterminate dendritic cell tumor/histiocytosis (IND-DCT) is a rare entity derived from indeterminate DC, the alleged precursors of Langerhans cells [1]; diagnosis is based on the expression of CD1a and S100, absence of Birbeck granules, and negativity for CD207. Four cases were diagnosed as IND-DCT on the basis of immunophenotype; two presented with cutaneous lesions (#156 by Dr. Lewis and #200 by Dr. Gindin), one involved a lymph node (#345 by Dr. Mendoza), and another the spleen (#339 by

Virchows Arch (2017) 471:467–489

475

Fig. 3 Langerhans cell histiocytosis. An unusual pure intrasinusoidal LCH mimicking reactive sinus histiocytosis (a and b, case #370, by Dr. Kuzu), showing strong expression of CD1a (c). Association of LCH with classical Hodgkin lymphoma (d, case#268, by Dr. Zhang) and with histiocytic proliferation containing numerous osteoclast-like giant cells (e, case #177, by Dr. Swerdlow; LC component not shown). Immunostains for BRAF V600E-mutated protein label only LCH cells (d, inset; f, case #177, right side), whereas Hodgkin cells (d, inset) and the histiocytic proliferation (f, left side) are negative

Dr. Aralica). On morphology, tumor cells were medium-large with abundant cytoplasm and showed mild (#200 and #345) to significant (#156 and #339) atypia with numerous mitoses, thus mimicking LCH and LCS/HS, respectively. Distinction form HS was particularly difficult in the splenic lesion, where tumor cells exhibited an intrasinusoidal growth pattern. Two cases of INDDCT were associated with other hematological neoplasias, including CMML and small B cell lymphoproliferation in case #345 (Fig. 5a, b) and splenic B cell lymphoma (case #339) (Table 2). Case #216 (Dr. Kansal) submitted as IND-DCT was defined by the panel as BDC sarcoma, NOS, with blastoid features^. It presented as an asymptomatic diffuse papulonodular skin eruptions in an 80-year-old man that rapidly disseminated leading to patient exitus in 6 months. Skin biopsy showed small–medium highly proliferating atypical cells with blastic features, expressing S100 and CD1a, lacking CD207 and Birbeck granules; they were positive for CD4,

CD11c, CD68/KP1, lysozyme, and TCL1, whereas other plasmacytoid dendritic cell markers (e.g., CD123, CD303, CD2AP) were negative. NCOA2-ETV3 gene fusion was recently identified in three cases of IND-DCT, while 11 LCH were negative [25], thus supporting the distinction between the two DC neoplasms. However, similarly to a previous report [26], IND-DCT case #345 showed BRAF V600E mutation, indicating that the boundary between LCH and IND-DCT may be subtle.

Interdigitating dendritic cell sarcoma Interdigitating dendritic cell sarcoma (IDCS) is a rare and aggressive disease made of atypical cells expressing S100 protein, but lacking CD1a and CD207. It commonly occurs in the lymph nodes, more rarely in extranodal sites [1]. Diagnosis is always made after exclusion of metastases and

476

Virchows Arch (2017) 471:467–489

Fig. 4 Langerhans cell histiocytosis and sarcoma. Figures a and b are from the same skin biopsy (case #368, by Dr. Feichtinger) showing areas with cytological features of LCH and LCS, respectively. CD207 immunostain in the corresponding areas is shown in c and d. LCS in a patient with previous T-ALL (e and f) (case #269, by Dr. Wright), showing strong expression of CD207 (f) and co-expression of CD1a (brown) and CD34 (blue) in same tumor cells (f, inset) (no lymphoblasts were present in the biopsy)

other DC neoplasms, especially melanoma and FDC sarcoma. Pathological and clinical features of IDCS were exemplified by case #159 (Dr. Metha), occurring in a 20-year-old male with multiple nodal and liver lesions, bone marrow involvement, and rapid progression (Fig. 5c). Markedly atypical tumor epithelioid cells effaced the lymph node parenchyma, diffusely expressed S100 (Fig. 5d), and were negative for CD1a, CD207 as other histiocytic, FDC, and epithelial and melanocytic markers. One of the three IDCS submitted cases showed nuclear positivity for SOX10 (#196, by Dr. Belyaeva); despite there are reports of alleged IDCS expressing this antigen [27, 28], much caution should be taken before providing this diagnosis in cases of malignant SOX10positive spindle cell neoplasms. In case #274, IDCS occurred in a patient with CLL and the two neoplasms were closely intermingled within the lymph node. Panel’s phenotypic criteria for defining IDCS are intense and diffuse expression of S100, negativity of CD1a and CD207, while variable

expression of CD68, CD163, CD45, and lysozyme is accepted. Case #299 (Dr. Hoeller) was submitted to the workshop as IDCS, but, on the basis of only focal reactivity for S100 protein and the expressionofCD68RandCD163,wasclassifiedasBDendriticcell sarcoma,nototherwisespecified^.Dataonthemutationallandscape ofIDCS are very limited;non recurrentgenetic anomalieswere detectedbyarrayCGHin3/4cases[29]andBRAFV600Emutationhas been reported in a single case [26], but not in others [30]; workshop cases#159and#274weretestedandresultednegative(Table3).

Follicular dendritic cell sarcoma Follicular dendritic cell sarcoma (FDCS) is a rare neoplasm consisting of cells with the morphology and phenotype of FDCs [1, 4, 5, 31–34]. FDCs are stromal-derived elements normally found in lymphoid follicles and can express antigens related to bone-marrow stroma [1, 4, 5, 31–34].

Virchows Arch (2017) 471:467–489

477

Fig. 5 Indeterminate dendritic cell tumor and interdigitating dendritic cell sarcoma. Figures a and b are from a case of INDDCT (case #345, by Dr. Mendoza) and show a lymph node involved by nodules (a) composed by bland cells with abundant eosinophilic cytoplasm (b); these cells expressed S100 (c, left) and CD1a (c, right), while CD207 was negative (not shown); note the unusual strong expression of BRAF V600Emutated protein in this tumor (d). Figures e and f are from a case of IDCS (case #159, by Dr. Mehta) that completely substitute the nodal parenchyma and is composed by epithelioid and spindle atypical cells with strong expression of S100 (f)

FDCS occurs at any age, although an adult predominance is observed [1, 4–6]. There is no sex predilection except for the inflammatory pseudotumor-like variant, most commonly affecting females and usually associated with EBV infection [1, 4, 5]. FDCS can develop in the setting of hyaline vascular Castleman disease (CD), where dysplastic FDCs are typically observed [4, 5, 31]. FDCS presents with lymph node disease in 31% of cases, extranodal disease in 58%, and both in about 10% [33]. At microscopic examination, FDCS consists of spindledovoid cells, forming fascicles, storiform arrays, whorls (at times with a 360° meningioma-like pattern), diffuse sheets, or vague nodules. Rare cases may also show jigsaw puzzlelike lobulation and perivascular spaces, mimicking thymoma or carcinoma showing thymus-like element (CASTLE) [1, 4–6, 31, 34]. A folliculotropic B cell-rich variant of FDCS has been reported [35].

Tumor cells have indistinct borders and contain a moderate amount of eosinophilic cytoplasm, and their nuclei are oval or elongated, with finely dispersed chromatin, small but distinct nucleoli, and a delicate nuclear membrane. Multinucleated giant cells and nuclear pseudoinclusions are frequently encountered. Frank cytological atypia is found in some cases with high mitotic rates (>30 mitotic figures per 10 highpower fields vs. 0–10 in the ordinary cases), atypical mitoses, and coagulative necrosis. The tumor is slightly infiltrated by small lymphocytes (B, T, or mixed). On immunophenotyping [1, 4–6, 31, 34], neoplastic cells express one or more of the FDC-associated markers including CD21, CD23, CD35, CXCL13 [36], clusterin, desmoplakin, podoplanin, claudin 4, vimentin, fascin, and EGFR (Table 4). Variably, they turn positive for EMA, protein S100, and CD68, while negative for CD1a, lysozyme, MPO, CD34, CD3, CD79a, and HMB45. Aberrant phenotypic expressions

3 5 5 5w

5

5 5 5

3 5 5 4 5

15/15 100%

116

141 164 185 194

210 248

259 283 316 334 346

356 361 366 382 387

+Cases/tot %

6/6 100%

5

4 5

5

5 5

D240

16/18 89%

4 5 5 0 5

5 5 0 5 5

5 5

5 5 5 5

5

5

CD21

7/9 78%

4

5 5 5 5

5

0 5

0

CD35

16/18 89%

2 5 5 4 5

5 5 5w 5 5

5 2

5w 5 0 0

5

5

CXCL13

11/15 73%

4 5 5 0

5 0 0 5 4

5

5 5 5 0

5

CD23

6/17 35%

0 0 0 0 0

1 2 5 0 4

0 0

0 2 0 3w

0

CD30

1/5 20%

0 0 5 0

0

CD31

1/6 17%

0

0

5 0

0

0

EMA

4/8 50%

5 0

3

0 0 0

3

5

Lys

4/7 57%

0

5 0 0

5w

5

5

CD4

5/13 38%

0 0 5 5

0

5w 0 0

0

5 3 (5)

0

0

CD68

1/5 20%

5

0

0 0

0

CD 163

2/15 13%

0 0 0 0 0

0

0 0 0

0 0

0 3

5

0

S100

0/7 0%

0

0

0

0

0

0

0

CD1a

2/8 25%

0 0 0 4

0

0

5

0

CD45

0 = 0% positive cells, 1 = 1–4%, 2 = 5–24%, 3 = 25–49%, 4 = 50–74%, 5 = 75–100%, w weak stain, CLUST clusterin, D240 podoplanin, Lys lysozyme

Clone for CD68 not specified except for case #194 scored 3 with PGM1 and 5 with KP1 and case #248 stained with clone KP1. When empty, immunostain not done

5

5

108

CLUST

Immunophenotypic findings in the 18 cases of FDCS

Case

Table 4

10/10 100%

5 5

5 5

5

5 5

5 5

5

PDL1

Average 25

50 15 60 10 5

70

5 (75 in met) 20

5 30

5

35

20

Ki-67 (%)

0/9 0%

0 0

0

0

0 0

0 0

0

EBER

478 Virchows Arch (2017) 471:467–489

Virchows Arch (2017) 471:467–489

can be encountered. The Ki-67 marking usually ranges from 1 to 25%, being higher in cases with frank atypia. Ultrastructurally, tumoral elements display long processes connected by scattered desmosomes. In one study, three of eight cases of FDCS showed clonal rearrangements of the immunoglobulin genes, in the absence of known associated lymphoma [37]. BRAF V600E mutation has at times been found [30]. Recently, four reports have shed new light on the pathobiology of FDCS [38–41]. A targeted sequencing study revealed recurrent loss-of-function of tumor suppressor genes involved in the negative regulation of NF-κB (38%) and cell cycle (31%) [38], including NFKBIA, CYLD, CDKN2A, and RB1. Focal copy number gain at 9p24 causing overexpression of CD274 (PD-L1) and PDCD1LG2 (PD-L2) was noted in three cases, which represents a well-known mechanism of immune evasion in cancer. Another study based on microRNA (miRNA) profiling of 31 FDCSs identified two subgroups with high and low miRNA expression levels, respectively [39]. The former appeared closer to fibroblasts and myopericytomas, whereas the latter to FDCs from CD. High miRNA-expressing group presented a tendency to a shorter overall survival and more frequent podoplanin expression. Laginestra et al. analyzed the transcriptome of 29 FDCSs and compared it with that of other mesenchymal tumors (MTs), microdissected CD FDCs, and normal fibroblasts [40]. The study demonstrated the transcriptional relationship of FDCSs with nonmalignant FDCs and their distinction from other MTs and fibroblasts. Furthermore, it provided evidence of a peculiar immunological microenvironment enriched in TFH and TREG populations, with special reference to the inhibitory immune receptor PD-1 and its ligands PD-L1 and PDL2. By a whole transcriptome study, Lorenzi et al. identified two novel FDC markers, the FDC secreted protein and serglycin, which integrate the above reported immunohistochemical panel [41] (Table 4). Most patients are treated by surgical excision followed or not by radiotherapy and/or chemotherapy [1, 4, 5]. Local recurrences and metastases are recorded in 50 and 25% of patients, respectively [1, 4, 5]. At least 10–20% of patients ultimately die of FDCS [1, 34, 38, 42]. Cases showing significant cytological atypia, extensive coagulative necrosis, high proliferation, size greater than 6 cm, and intra-abdominal location can run a rapidly fatal course [1, 4–6]. A lymphoma-like therapy is adopted in the latter cases, although no reference protocol does exist [4, 5]. Nineteen cases were submitted to the workshop: 18 were classified as FDCS and one represented an example of TAFRO syndrome (thrombocytopenia, ascites/anasarca, microcytic anemia, myelofibrosis, renal dysfunction, and organomegaly) [43]. The morphologic spectrum of FDCS was recapitulated in cases #116 (by Dr. Said), #248 (by Dr. Zecchini Barrese),

479

#361 (by Dr. Kaygusuz), and #387 (by Dr. Greiner) (Fig. 6a–d). Cases #194 (by Dr. Bhagat), #210 (by Dr. Riley), #283 (by Dr. Wotherspoon), and #346 (by Dr. Nakashima) developed within the context of CD by showing the transition from typical hyaline vascular CD to areas with FDC dysplasia to frank FDCS (Fig. 6e). Unusual morphologic features were seen in six cases. Case #164 (by Dr. Lorenzi) displayed an unprecedently reported Bangiomatoid^ architecture, which had originally led to a misdiagnosis of angiosarcoma. The staining for CD31 was negative, while seven FDC-associated markers were expressed (Fig. 7a). Case #356 (by Dr. Pugh) corresponded to the description of the inflammatory pseudotumor-like variant of FDCS but was EBV-negative (Fig. 7b). Case #382 (by Dr. Dartigues) occurred in patient with a previous history of indolent B cell lymphoma, who achieved partial remission following two lines of chemotherapy. A huge mesenteric mass developed, which was completely removed and diagnosed as FDCS. On microscopic examination, the latter showed a biphasic pattern with micronodules, consisting of spindle cells expressing the FDC-associated markers clusterin and CXCL13, surrounded by a fusiform component of indeterminate origin, which was negative for FDC markers and instead positive for CD68, CD163, and CD11c (Fig. 7c, d). A biphasic pattern was also observed in case #346, corresponding to a submandibular mass. The specimen showed some nodules consisting of spindle cells (CD21+, CD23+, CD30+), surrounded by an epithelioid cell component (CD21+, CD23−, CD30−, lysozyme+) (Fig. 8a). The partial CD30 positivity, encountered in other five cases (#164; #194; #259 by Dr. Sadigh; #283; #316 by Dr. Milowich), might represent the target for an anti-CD30 immunoconjugate [44]. Moreover, it highlights the immunophenotypical variability of FDCS, with frequent defectivity of some of the FDC-associated markers and possible phenotypic aberrations (Table 4) (Fig. 8b, c). This was exemplified by the expression of CD4 detected in four cases (#108 by Dr. Malysz; #116; #141 by Dr. Raymond; #259) and positivity for CD31 and CD56 in one case each (#185 by Dr. Morlote; #334 by Dr. Bozkurt). High Ki-67 expression was found in five cases (#141; #259; #346; #356; and #366 by Dr. Feichtinger) (Fig. 8d). In case #259, presenting as a neck mass, the high proliferation rate corresponded to a focal area (Ki-67: 75 vs. 5% of the rest), also characterized by a trabecular growth pattern and loss of CD21 and CD35. Identical histologic and phenotypic features of the highly proliferative area characterized the lung metastasis that occurred 2 years later. Interestingly, despite partial tumor resection, the patient was alive and well 7 years after disease presentation. In two of the remaining cases with high Ki-67 marking (35–70%), no clinical data are available, while the third (#366) was lost at follow-up with disease progression. The latter patient presented with a large mass extensively involving the pharyngeal region and accompanied by bilateral

480 Fig. 6 Follicular dendritic cell sarcoma. Case #361 (by Dr. Kaygusuz) displayed classical FDCS growth pattern, with storiform and/or whorled bundles of spindle cells (a); at higher magnification, neoplastic cells showed mild atypia with elongated nuclei (b). Case #141 (by Dr. Raymond) was characterized by eosinophilic nuclear inclusions (c). Increased cytological atypia and Banaplastic^ features were observed in case #316 (d, by Dr. Milowich). Case #283: concurrent involvement by FDCS (e, by Dr. Wotherspoon) and Castleman’s disease (e, inset)

Fig. 7 Follicular dendritic cell sarcoma. Uncommon FDCS growth patterns: angiosarcomalike in case #164 (a, by Dr. Lorenzi) and pseudoinflammatory type in case #356 (b, by Dr. Pugh). In case #382 (by Dr. Dartigues), a micronodular FDCS component CXCL13+ (c) was associated with a spindle cell indeterminate proliferation (d: CD163 staining)

Virchows Arch (2017) 471:467–489

Virchows Arch (2017) 471:467–489

481

Fig. 8 Follicular dendritic cell sarcoma. Strong CD30 expression in case #346 (a, by Dr. Nakashima). An incomplete FDC phenotype was detected in case #283 (by Dr. Wotherspoon) with positivity for CD21 (b) and lack of CD23 (c). High Ki-67 proliferation index in case #356; please note that in this case with a pseudoinflammatory appearance, neoplastic cells were scanty; however, about a half of them were proliferating (d, by Dr. Pugh). Diffuse and strong membrane reaction for PD1ligand in case #108 (e, by Dr. Malysz)

cervical lymph adenopathy; he underwent radiotherapy with poor response rate. In keeping with recent data provided by Laginestra et al. [40], all ten FDCS cases tested for PDL1 were strongly positive (Fig. 8e), supporting the potential application of immune checkpoint inhibitors for FDCS treatment [13, 14]. No conclusion can be drawn from the submitted cases concerning the best therapeutic option due to the heterogeneity of treatments used. Overall, seven out of eight patients with follow-up data were alive and well from 1 to 11 years after the diagnosis.

Erdheim-Chester disease, Rosai-Dorfman disease, and ALK+ histiocytosis Erdheim-Chester disease (ECD) shares molecular features with LCH, [45, 46] but represents a distinct disease, characterized by clonal systemic proliferation of histiocytes often with xanthomatous features including Touton-like cells and fibrosis. Any organ, but especially long bones, cardiovascular system, and retroperitoneum, can be involved by ECD [1, 47]. Diagnosis relies on strict correlation between clinical features, imaging, and histology [47, 48]. The workshop received two cases of ECD, one involving skin and soft tissues in a 26-yearold female (case #267, Dr. Durham), one occurring in multiple

sites in a 44-year-old male (case #257, Dr. Theate), where a biopsy of the omentum revealed foci of foamy histiocytes with rare Touton-like cells and tiny bands of fibrosis (Fig. 9a). CD68 and CD163 were strongly positive (Fig. 9b), while S100 and CD1a were not expressed. BRAF V600E-mutated protein was expressed by the histiocytes (Fig. 9b), highlighting the diagnostic usefulness of immunohistochemistry to exclude reactive causes of histiocytic fat infiltration [48]. Five cases of Rosai-Dorfman disease (RDD) were submitted to the workshop, and all of them were from extranodal sites (minor salivary glands, #121 Dr. Margolskee; subcutis, #179 Dr. Harmon, Fig. 9c; skull, #188 Dr. Shet, gastrointestinal tract, #215 Dr. Hernandez, and heart wall, #340 Dr. Robin). In all cases but one (#340), the disease was multifocal or multicentric. In two, the clinical presentation was exceptionally uncommon, with multiple gastrointestinal polyps in a 60-year-old female (#215) or an isolated heart mass in a 3year-old girl (#340). According to their extranodal location, the typical histological features of RDD were not straightforwardly recognizable, especially in cases #188 and #215, with spindly RDD cells, prominent fibrosis, and less obvious emperipolesis (Fig. 9d). Immunostains were homogeneously characterized by the expression of S100 in a significant but variable percentage of large cells, together with histiocytic markers (factor XIIIa, CD68, CD163); CD1a and CD207 were negative. Different

482

Virchows Arch (2017) 471:467–489

Fig. 9 Erdheim-Chester disease. Case #257 (by Dr. Theate) shows foamy histiocytes and rare Touton-like cells associated with bland fibrosis (a); tumor cells are strongly positive for CD163 (b, left) and part of them for BRAF V600E-mutated protein (b, right). Rosai-Dorfman disease. Figure c shows typical cytological features in a subcutaneous tumor (c, case #179, by Dr. Harmon), while figure d displays a spindly morphology in a rectal polyp (d, case#215, by Dr. Hernandez). ALK-positive histiocytosis. Kidney involvement by large cells (e) showing marked similarities to RDD cells, including remarkable emperipolesis and reactivity for S100 (f, left); strong cytoplasmic expression of ALK protein is shown in figure f, right

treatments have been used in RDD without defined protocols, with heterogeneous clinical responses [2]. In case #188, an 8year-old boy had a disseminated disease unresponsive to multiple regimens, including vinblastine, prednisolone, etoposide, cyclophosphamide, hydroxyl-daunorubicin, and imatinib. Although RDD has been considered a reactive process [49], the recent demonstration of somatic mutation in some cases, including KRAS K117N mutation found in case #121 [50], challenged this hypothesis [12, 46]. Furthermore, RDD may occur in individuals with germline mutations of the SLC29A3 gene (Faisalabad histiocytosis/familial RosaiDorfman disease) [51] and in autoimmune lymphoproliferative syndrome type Ia associated with TNFRSF6 mutations [52]. Interestingly, histological and immunohistochemical (S100+) features of RDD were recognizable in the exceptional case of ALK-positive histiocytosis involving the kidney and the liver (by Dr. Hintzke #371). In analogy with other similar reported cases [53], this disease occurred in a severely ill

infant, with multicentric accumulation of ALK+ and S100+ histiocytes (Fig. 9e, f), expressing other histiocytic markers but not CD1a nor CD207. In case #371, ALK reactivity was membranous and cytoplasmic (Fig. 9f), and ALK translocation was detected using a break apart FISH probe. Immunostain for ALK was performed by the panel in all RDD workshop cases and resulted negative, in keeping with previous data [53].

Blastic plasmacytoid dendritic cell neoplasm Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare and aggressive hematologic neoplasm with striking cutaneous tropism and frequent bone marrow involvement and systemic spread [1]. BPDCN mainly arises in adults (median age at diagnosis of 65.0 years) [54], but about 10% of cases have been reported in young individuals [54], as illustrated by

Virchows Arch (2017) 471:467–489

case #204 (by Dr. Yiğit) and #182 (by Dr. Balagué), occurring in females of 11 and 17 years, respectively. The four cases of BPDCN submitted to the workshop presented with skin lesions from the beginning, which had arisen from four to 12 months before work-up. Skin biopsies showed a non-epidermotropic diffuse infiltrate of medium-sized blasts (Fig. 10a), with consistent expression of PDC markers (partially performed by the panel), such as CD123 (4/4), TCL1 (4/4), BDCA2/ CD303 (3/4) (Fig. 10b), and CD2AP (4/4); CD56 was positive in all cases, while CD4 was negative in one (#204), as occurs in about 5% of BPDCN cases [54]. CD68 was expressed as single paranuclear dots in three cases, but was unusually diffuse in one (#182) (Fig. 10b). Finally, the diagnostic usefulness of MXA (alpha-interferon induced myxovirus resistance gene A) [55] was confirmed by the panel demonstration of positivity of all cases (Fig. 10b). Overall, the results obtained in these four cases of BPDCN are in keeping with literature data regarding criteria to distinguish BPDCN from acute myelogenous and lymphoblastic leukemia, that requires positivity of CD4 and/or CD56 and at least two of the 4 PDC markers (CD123, CD303, TCL1, MXA), with negativity for CD3, CD20, myeloperoxidase, and lysozyme [54–57]. Recently, the high sensitivity and specificity of an antibody recognizing the master PDC transcription factor TCF4/E2–2 has been shown [58]. About 6% of cases of BPDCN are associated with another myeloid neoplasm, particularly myelodysplasia and acute myeloid leukemia that may precede, concur, or follow BPDCN [57]. In these cases, myeloid leukemic cells may share with BPDCN positivity for CD4 and CD56, as well as TCL1 and CD123. Case #100 (Dr. Cogliatti) well illustrates such a condition of a primary cutaneous BPDCN with full-blown immunophenotype, whose bone marrow was involved by CD4+/CD56+/CD13+/CD64+/myeloperoxidase+ blasts representing acute myeloid leukemia. The patient, a 73-year-old male, did not respond to a variety of chemotherapeutic protocols (CHOEP, GMALL elderly protocol, CNS MTX prophylaxis) and rapidly deteriorated with tumor dissemination and washout of blastic plasmacytoid dendritic tumor cells. The mutational profile of BPDCN analyzed by massive parallel sequencing identified TET2 as the most commonly mutated gene [59–61]. In addition, novel deleterious mutations of the transcription factors IKZF3 and ZEB2, never observed in other hematologic malignancies, were found in 12– 16% of cases [61]; on the other hand, among mutations typically found in myeloid neoplasms, BPDCN displays very rarely some of them (i.e., JAK2 and FLT3), but rather frequently others (i.e., ASXL1, NPM1, and H/K/N RAS)

483

[60–62]. In case #204, targeted sequencing evaluating 405 different cancer-related genes identified inactivating deleterious mutation of PBRM1, commonly seen in clear cell renal carcinoma [63], resulting in a severely truncated protein. The presence of mutation in one allele and the probably gross deletion of the other are consistent with a tumor suppressor function of PBRM1, a component of the SWI/SNF chromatin remodeling complex. Interestingly, in one study, half BPDCN cases had mutations affecting either the DNA methylation or chromatin remodeling pathways [61]. BCL2 gene and protein have been found to be consistently overexpressed in BPDCN [58, 64], as were in all workshop cases (Fig. 10b); similar result was obtained by the panel in a separate series of 16 BPDCN, where no BCL2 anomalies were detected using break apart FISH probes, thus suggesting that BCL2 overexpression likely depends on gene mutations or abnormal gene regulation. BCL2 expression was investigated in PDC occurring in various inflammatory conditions (e.g., reactive lymphadenitis, Kikuchi’s lymphadenitis, cutaneous lupus erythematosus), as well as in six cases of mature PDC proliferation associated with chronic myelomonocytic leukemia [1, 54], all resulting negative (Fig. 10c, d). The overexpression of BCL2 is therefore a hallmark of PDC in BPDCN. No significant conclusions can be made from the four workshop cases about clinical evolution, since one patient died of disease, and in the other three, the follow-up was too short. There is no consensus regarding the optimal treatment modality of BPDCN. In patients in first complete remission, allogeneic transplantation has been advocated as the best way to obtain long-term survival [65, 66]. Treatment options could be explored on the basis of translational studies in vitro and in xenografts based on novel molecular findings, including agents targeting the NFkB pathway aberrantly activated in BPDCN [64], the BCL2 inhibitor venetoclax [67], and bromodomain and extraterminal domain (BET) protein inhibitors, disrupting a TCF4/E2-2 dependent BPDCN-specific transcriptional network [58].

Immunophenotypic markers used in the diagnosis of tumors derived from histiocytes/macrophages (MAC) and hematopoietic dendritic cells (HDC) Identification and subclassification of the neoplasms derived from histiocytes and dendritic cells are based on several immunophenotypic markers, whose sensitivity and specificity are quite variable, as detailed in Table 1. It should be noted that the majority of H/M markers are also expressed in myeloid sarcoma of monocytic origin. Histiocytes and dendritic cells are efficient antigen presenting cells, and their expression of major histocompatibility class II molecules is pivotal in this process [68]. On the basis of a recent report showing that HLA-DR displays a distinctive

484

Virchows Arch (2017) 471:467–489

Fig. 10 Blastic plasmacytoid dendritic cell neoplasm. A classic example involving the skin composed of medium-size blasts (a, case #182, by Dr. Balagué), strongly positive for CD303/ BDCA2 (b, upper left) and MXA (b, upper right); in this case, tumor cells displayed unusual diffuse reactivity for CD68 (b, lower left). Note also the strong expression of BCL2 typically found in BPDCN (b, lower right), which contrasts with the negativity regularly found in PDC aggregates in reactive lymph nodes (c) and in proliferations of mature PDC associated with chronic myelomonocytic leukemia (d) (c and d are from of one of the authors’ files (FF))

and exclusive cytoplasmic globular stain pattern in LCH [69], the panel performed HLA-DR immunostain in 11 reactive lymph nodes and in 33 histiocytic and dendritic cell neoplasms, including FDC-S (13 cases), HS (4), RDD (4), LCH (5), 2 cases each of LCS, IDCS, and IND-DCT, and 1 DCSNOS. In reactive lymph nodes, interdigitating dendritic cells showed intense membrane positivity (Fig. 11a), tingible body macrophages expressed HLA-DR in the cytoplasm, while sinus macrophages were negative or fairly positive (Fig. 11a). Highly intense, exclusively cytoplasmic, globular or combined globular, and granular reactivity was confirmed in all five LCH (Fig. 11b), as well as in LCS (Fig. 11c), IND-DCT (Fig. 11d), and IDCS (Fig. 11e) cases. In contrast, this stain pattern was never detected in the remaining neoplasms, where tumor cells were either completely negative or showed a combination of membranous and/or non-globular cytoplasmic reactivity. Moreover, HLA-DR was completely negative in RDD (Fig. 11f), in keeping with previous observations [70]. These results indicate that HLA-DR stain pattern may be helpful in distinguishing neoplastic from reactive DC proliferations, and distinguish LCH, LCS, IND-DCT, and IDCS from other tumoral categories. Since histiocytic and dendritic cell tumors may present the BRAF V600E mutation which may be used for treatment selection, the anti-BRAF V600E VE1 antibody is helpful at least for screening purposes [71]. Good concordance between immunohistochemistry and molecular analysis has been reported in several studies [72–74]. Nineteen cases submitted to the workshop (5 HS, 5 LCH, 1 LCS, 3 IND-DCT, 1 DCT-NOS,

2 IDCS, 1 RDD, 1 ECD) were stained for VE1 either by the authors and/or by the panel, and reactivity was found in six of them (1 HS, 3 LCH, 1 IND-DCT, 1 ECD) (Table 3). Molecular analysis using Sanger sequencing, pyrosequencing, or NGS were performed in ten of these cases, and the immunohistochemical results (6 positive, 4 negative) were confirmed in all of them. Interestingly, in three positive cases (2 LCH, #268 and #347; 1 IND-DCT #345), Sanger sequencing performed in two different laboratories failed to detect the mutation, which on the contrary was identified by NGS (4, 2, and 16% allele frequency, respectively, in cases #268, #347, and #345). These results are in keeping with the recommendation to use more sensitive techniques to detect mutations in these neoplasms, given their low allelic frequency [2, 45, 46].

HS and DC neoplasms association with lymphoid malignancies The association of histiocytic and dendritic cell neoplasms with solid tumors, acute leukemias, or lymphomas is a wellrecognized event [8, 75–82]. Classical Hodgkin’s and nonHodgkin’s lymphomas (especially B cell follicular lymphoma, CLL/small lymphocytic lymphoma, diffuse large B cell lymphoma, and B and T lymphoblastic lymphoma/leukemia) may occur at the same or at different anatomical sites, synchronously or metachronously, when the lymphoma generally occurs first [75, 76, 81–86].

Virchows Arch (2017) 471:467–489

485

Fig. 11 HLA-DR expression in histiocytic and dendritic cell neoplasms. Interdigitating dendritic cells of the paracortex in a reactive lymph node display strong membranous reactivity for HLA-DR (a, right), while sinus histiocytes are negative (a, left). Tumor cells in cases of Langerhans cell histiocytosis (b, case #370, by Dr. Kuzu), Langerhans cell sarcoma (c, case #122, by Dr. Gao), indeterminate dendritic cell tumor (d, case #345, by Dr. Mendoza), and interdigitating dendritic cell sarcomas (e, case #274, by Dr. Diepstra) display a globular or combined globular and granular cytoplasmic reactivity (details in insets). In histiocytes of RosaiDorfman diseases, HLA-DR was regularly negative (f, case #179, by Dr. Harmon)

Excluding cases in which accumulation of histiocytes or dendritic cells (e.g., Langerhans cells) simply represent an exaggerated reaction [79], the occurrence of two distinct neoplasms has been considered to depend from genetic predisposition or treatment effects. Moreover, in most studies, a molecular identity and clonal relationship between the two processes has been proven [8, 80, 81], explained either by the derivation of the two neoplasms from a common pluripotent precursor [81] or by the so called Btransdifferentiation.^ This process is equivalent to the in vitro reprogramming of B lymphocytes into macrophages by few molecular hits [8, 87], either directly or through subsequent steps of dedifferentiation and redifferentiation [81, 88]. Fourteen cases of the workshop revealed this association (Table 2), HS being the mostly represented tumor (five cases), followed by LCH (four), LCS (two), and IND-DCT (two); lymphomas were mainly of low-grade B cell subtype (FL and CLL occurring in four cases each), and in the majority

of cases (10/14, 71.4%), they occurred first (from several months to more than 20 years), while coexisted in the same biopsy in three cases (3/14, 21.4%). In a single case (case #362), LCH occurred 4 years before the diagnosis of CLL and shared the same BCR rearrangement. In 11/14 cases (78.6%), a molecular relationship between the two neoplasms could be demonstrated, including case #307, where FL and subsequent HS shared double BCL2 and MYC translocations, and case #306 with IGH-CCND1 fusion detected in myeloma and HS (Table 2). Of notice, despite in case #122 the two neoplasms shared the IGH/BCL2 translocation, IG clonal rearrangement could be demonstrated only in FL. Similarly, BRAF V600E mutation was detected by immunohistochemistry only in two histiocytic/dendritic cell neoplasms (#268 and #345), but not in the associated lymphoma occurring in the same lymph node. In these cases, the BRAF mutation might have occurred very late in the dendritic cell population, according to the hypothesis by

486

Berres et al. [15, 20]. Case #363, on the other hand, showed transformation of a BRAF V600E-mutated hairy cell leukemia in an histiocytic sarcoma carrying the same and additional mutations [89]. Clonal immunoglobulin genes rearrangement and/or IGH translocations in histiocytic and dendritic cell neoplasm may occur in patients lacking clinical or histological evidence of lymphomas [37, 81, 90]. Interestingly, this phenomenon was reported in two nonmalignant histiocytic proliferations from the workshop. In case #179, IGH clonality was detected in multifocal RDD with prominent plasma cell infiltrate, initially misdiagnosed as plasma cell neoplasm. In case #183 (by Dr. Patel), a histiocytic neoplasm with features of reticulohistiocytoma showed clonal IGH rearrangement and IGH translocation by FISH. Although the existence of a regressed lymphoid neoplasm could not be excluded, Bpseudoclonality^ due to bystander lymphocytes should always be considered, especially in these enigmatic cases.

Virchows Arch (2017) 471:467–489 Informed consent Informed consent in this study is not applicable, since all cases were provided by different participants to the workshop and all data were completely anonymized.

References 1.

2.

3.

4.

Molecular landscape of dendritic and histiocytic neoplasms Recent studies on dendritic cell and histiocytic neoplasms demonstrated that disorders significantly diverse on clinical, morphological, and phenotypic features share molecular alterations, some of which may have therapeutic usefulness [2, 46]. Table 3 reports the results of molecular studies on cases from the workshop performed by the authors and/or the panel, using various modalities. As expected, the most common molecular anomalies involved BRAF and the genes related to the RAS-RAF-ERK pathway, which were observed in ten cases of different histotype (Table 3). Of notice, efficacy of mutation-targeted therapy was reported in two cases of HS from the workshop. In case #243, the BRAF V600E-mutated tumor was treated with vemurafenib after failure of intensive chemotherapy (EPOCH), with significant decrease in tumor size and metabolic activity. MEK inhibitor as single agent was given in case #372 carrying MAP2K1 mutation and induced complete response in a highly compromised patient with multifocal disease. Overall, these data further support the role played by the RAS-RAF-MEK-ERK pathway in tumorigenesis of histiocytic neoplasms, pointing out the need of molecular analysis in order to identify potential therapeutic target. Acknowledgements The study was supported by Associazione Italiana per la Ricerca sul Cancro (AIRC) grants 5x1000 10007 and IG 15762 (to S.A.P.) and by Fondazione Golgi, Brescia (to F.F.).

5.

6.

7.

8.

9.

10.

11.

Compliance with ethical standards Conflict of interest The authors declare that they have no conflicts of interest.

12.

Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW (2008) WHO classification of tumours of haematopoietic and lymphoid tissues. World Health Organization Classification of Tumorus, Lyon Emile JF, Abla O, Fraitag S, Horne A, Haroche J, Donadieu J, Requena-Caballero L, Jordan MB, Abdel-Wahab O, Allen CE, Charlotte F, Diamond EL, Egeler RM, Fischer A, Herrera JG, Henter JI, Janku F, Merad M, Picarsic J, Rodriguez-Galindo C, Rollins BJ, Tazi A, Vassallo R, Weiss LM (2016) Revised classification of histiocytoses and neoplasms of the macrophage-dendritic cell lineages. Blood 127:2672–2681 Swerdlow SH, Campo E, Pileri SA, Harris NL, Stein H, Siebert R, Advani R, Ghielmini M, Salles GA, Zelenetz AD, Jaffe ES (2016) The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood 127:2375–2390 Dalia S, Jaglal M, Chervenick P, Cualing H, Sokol L (2014) Clinicopathologic characteristics and outcomes of histiocytic and dendritic cell neoplasms: the moffitt cancer center experience over the last twenty five years. Cancers (Basel) 6:2275–2295 Dalia S, Shao H, Sagatys E, Cualing H, Sokol L (2014) Dendritic cell and histiocytic neoplasms: biology, diagnosis, and treatment. Cancer Control 21:290–300 Pileri SA, Grogan TM, Harris NL, Banks P, Campo E, Chan JK, Favera RD, Delsol G, De Wolf-Peeters C, Falini B, Gascoyne RD, Gaulard P, Gatter KC, Isaacson PG, Jaffe ES, Kluin P, Knowles DM, Mason DY, Mori S, Muller-Hermelink HK, Piris MA, Ralfkiaer E, Stein H, Su IJ, Warnke RA, Weiss LM (2002) Tumours of histiocytes and accessory dendritic cells: an immunohistochemical approach to classification from the International Lymphoma Study Group based on 61 cases. Histopathology 41: 1–29 Takahashi E, Nakamura S (2013) Histiocytic sarcoma: an updated literature review based on the 2008 WHO classification. J Clin Exp Hematop 53:1–8 Feldman AL, Arber DA, Pittaluga S, Martinez A, Burke JS, Raffeld M, Camos M, Warnke R, Jaffe ES (2008) Clonally related follicular lymphomas and histiocytic/dendritic cell sarcomas: evidence for transdifferentiation of the follicular lymphoma clone. Blood 111: 5433–5439 Wang E, Hutchinson CB, Huang Q, Sebastian S, Rehder C, Kanaly A, Moore J, Datto M (2010) Histiocytic sarcoma arising in indolent small B-cell lymphoma: report of two cases with molecular/genetic evidence suggestive of a ‘transdifferentiation’ during the clonal evolution. Leuk Lymphoma 51:802–812 Idbaih A, Mokhtari K, Emile JF, Galanaud D, Belaid H, de Bernard S, Benameur N, Barlog VC, Psimaras D, Donadieu J, Carpentier C, Martin-Duverneuil N, Haroche J, Feuvret L, Zahr N, Delattre JY, Hoang-Xuan K (2014) Dramatic response of a BRAF V600Emutated primary CNS histiocytic sarcoma to vemurafenib. Neurology 83:1478–1480 Xue K, Song J, Yang Y, Li Z, Wu C, Jin J, Li W (2016) PAX5 promotes pre-B cell proliferation by regulating the expression of pre-B cell receptor and its downstream signaling. Mol Immunol 73:1–9 Gatalica Z, Bilalovic N, Palazzo JP, Bender RP, Swensen J, Millis SZ, Vranic S, Von Hoff D, Arceci RJ (2015) Disseminated

Virchows Arch (2017) 471:467–489

13. 14. 15.

16.

17.

18.

19.

20.

21.

22. 23.

24.

25.

26.

27.

28.

histiocytoses biomarkers beyond BRAF V600E: frequent expression of PD-L1. Oncotarget 6:19819–19825 Balar AV, Weber JS (2017) PD-1 and PD-L1 antibodies in cancer: current status and future directions. Cancer Immunol Immunother Galanina N, Kline J, Bishop MR (2017) Emerging role of checkpoint blockade therapy in lymphoma. Ther Adv Hematol 8:81–90 Collin M, Bigley V, McClain KL, Allen CE (2015) Cell(s) of origin of Langerhans cell histiocytosis. Hematol Oncol Clin North Am 29: 825–838 Chakraborty R, Burke TM, Hampton OA, Zinn DJ, Lim KP, Abhyankar H, Scull B, Kumar V, Kakkar N, Wheeler DA, Roy A, Poulikakos PI, Merad M, McClain KL, Parsons DW, Allen CE (2016) Alternative genetic mechanisms of BRAF activation in Langerhans cell histiocytosis. Blood 128:2533–2537 Badalian-Very G, Vergilio JA, Degar BA, MacConaill LE, Brandner B, Calicchio ML, Kuo FC, Ligon AH, Stevenson KE, Kehoe SM, Garraway LA, Hahn WC, Meyerson M, Fleming MD, Rollins BJ (2010) Recurrent BRAF mutations in Langerhans cell histiocytosis. Blood 116:1919–1923 Nelson DS, van Halteren A, Quispel WT, van den Bos C, Bovee JV, Patel B, Badalian-Very G, van Hummelen P, Ducar M, Lin L, MacConaill LE, Egeler RM, Rollins BJ (2015) MAP2K1 and MAP3K1 mutations in Langerhans cell histiocytosis. Genes Chromosomes Cancer 54:361–368 Nelson DS, Quispel W, Badalian-Very G, van Halteren AG, van den Bos C, Bovee JV, Tian SY, Van Hummelen P, Ducar M, MacConaill LE, Egeler RM, Rollins BJ (2014) Somatic activating ARAF mutations in Langerhans cell histiocytosis. Blood 123: 3152–3155 Berres ML, Lim KP, Peters T, Price J, Takizawa H, Salmon H, Idoyaga J, Ruzo A, Lupo PJ, Hicks MJ, Shih A, Simko SJ, Abhyankar H, Chakraborty R, Leboeuf M, Beltrao M, Lira SA, Heym KM, Clausen BE, Bigley V, Collin M, Manz MG, McClain K, Merad M, Allen CE (2015) BRAF-V600E expression in precursor versus differentiated dendritic cells defines clinically distinct LCH risk groups. J Exp Med 212:281 Edelweiss M, Medeiros LJ, Suster S, Moran CA (2007) Lymph node involvement by Langerhans cell histiocytosis: a clinicopathologic and immunohistochemical study of 20 cases. Hum Pathol 38: 1463–1469 Bechan GI, Egeler RM, Arceci RJ (2006) Biology of Langerhans cells and Langerhans cell histiocytosis. Int Rev Cytol 254:1–43 Valladeau J, Ravel O, Dezutter-Dambuyant C, Moore K, Kleijmeer M, Liu Y, Duvert-Frances V, Vincent C, Schmitt D, Davoust J, Caux C, Lebecque S, Saeland S (2000) Langerin, a novel C-type lectin specific to Langerhans cells, is an endocytic receptor that induces the formation of Birbeck granules. Immunity 12:71–81 Verdijk P, Dijkman R, Plasmeijer EI, Mulder AA, Zoutman WH, Mieke Mommaas A, Tensen CP (2005) A lack of Birbeck granules in Langerhans cells is associated with a naturally occurring point mutation in the human Langerin gene. J Invest Dermatol 124:714– 717 Brown RA, Kwong BY, McCalmont TH, Ragsdale B, Ma L, Cheung C, Rieger KE, Arber DA, Kim J (2015) ETV3-NCOA2 in indeterminate cell histiocytosis: clonal translocation supports sui generis. Blood 126:2344–2345 O'Malley DP, Agrawal R, Grimm KE, Hummel J, Glazyrin A, Dim DC, Madhusudhana S, Weiss LM (2015) Evidence of BRAF V600E in indeterminate cell tumor and interdigitating dendritic cell sarcoma. Ann Diagn Pathol 19:113–116 Stowman AM, Mills SE, Wick MR (2016) Spindle cell melanoma and interdigitating dendritic cell sarcoma: do they represent the same process? Am J Surg Pathol 40:1270–1279 Magro CM, Olson LC, Nuovo G, Solomon GJ (2017) Primary cutaneous interdigitating dendritic cell sarcoma is a morphologic and phenotypic simulator of poorly differentiated metastatic

487 melanoma: a report of 2 cases and review of the literature. Ann Diagn Pathol 29. O'Malley DP, Zuckerberg L, Smith LB, Barry TS, Gunn S, Tam W, Orazi A, Kim YS, Weiss LM (2014) The genetics of interdigitating dendritic cell sarcoma share some changes with Langerhans cell histiocytosis in select cases. Ann Diagn Pathol 18:18–20 30. Go H, Jeon YK, Huh J, Choi SJ, Choi YD, Cha HJ, Kim HJ, Park G, Min S, Kim JE (2014) Frequent detection of BRAF (V600E) mutations in histiocytic and dendritic cell neoplasms. Histopathology 65:261–272 31. Facchetti F, Lorenzi L (2016) Follicular dendritic cells and related sarcoma. Semin Diagn Pathol 33:159–166 32. Perkins SM, Shinohara ET (2013) Interdigitating and follicular dendritic cell sarcomas: a SEER analysis. Am J Clin Oncol 36:395–398 33. Saygin C, Uzunaslan D, Ozguroglu M, Senocak M, Tuzuner N (2013) Dendritic cell sarcoma: a pooled analysis including 462 cases with presentation of our case series. Crit Rev Oncol Hematol 88:253–271 34. Ohtake H, Yamakawa M (2013) Interdigitating dendritic cell sarcoma and follicular dendritic cell sarcoma: histopathological findings for differential diagnosis. J Clin Exp Hematop 53:179–184 35. Lorenzi L, Lonardi S, Petrilli G, Tanda F, Bella M, Laurino L, Rossi G, Facchetti F (2012) Folliculocentric B-cell-rich follicular dendritic cells sarcoma: a hitherto unreported morphological variant mimicking lymphoproliferative disorders. Hum Pathol 43:209–215 36. Vermi W, Lonardi S, Bosisio D, Uguccioni M, Danelon G, Pileri S, Fletcher C, Sozzani S, Zorzi F, Arrigoni G, Doglioni C, Ponzoni M, Facchetti F (2008) Identification of CXCL13 as a new marker for follicular dendritic cell sarcoma. J Pathol 216:356–364 37. Chen W, Lau SK, Fong D, Wang J, Wang E, Arber DA, Weiss LM, Huang Q (2009) High frequency of clonal immunoglobulin receptor gene rearrangements in sporadic histiocytic/dendritic cell sarcomas. Am J Surg Pathol 33:863–873 38. Griffin GK, Sholl LM, Lindeman NI, Fletcher CD, Hornick JL (2016) Targeted genomic sequencing of follicular dendritic cell sarcoma reveals recurrent alterations in NF-kappaB regulatory genes. Mod Pathol 29:67–74 39. Hartmann S, Doring C, Agostinelli C, Portscher-Kim SJ, Lonardi S, Lorenzi L, Fuligni F, Martinez D, Mehta J, Borges A, Hackstein H, Kippenberger S, Piccaluga PP, Simonitsch-Klupp I, Cabecadas J, Campo E, Facchetti F, Pileri SA, Hansmann ML (2016) miRNA expression profiling divides follicular dendritic cell sarcomas into two groups, related to fibroblasts and myopericytomas or Castleman’s disease. Eur J Cancer 64:159–166 40. Laginestra MA, Tripodo C, Agostinelli C, Motta G, Hartmann S, Doring C, Rossi M, Melle F, Sapienza MR, Tabanelli V, Pileri A, Fuligni F, Gazzola A, Mannu C, Sagramoso CA, Lonardi S, Lorenzi L, Bacci F, Sabattini E, Borges A, Simonitsch-Klupp I, Cabecadas J, Campo E, Rosai J, Hansmann ML, Facchetti F, Pileri SA (2017) Distinctive histogenesis and immunological microenvironment based on transcriptional profiles of follicular dendritic cell sarcomas. Mol Cancer Res 41. Lorenzi L, Doring C, Rausch T, Benes V, Lonardi S, Bugatti M, Campo E, Cabecadas J, Simonitsch-Klupp I, Borges A, Mehta J, Agostinelli C, Pileri SA, Facchetti F, Hansmann ML, Hartmann S (2017) Identification of novel follicular dendritic cell sarcoma markers, FDCSP and SRGN, by whole transcriptome sequencing. Oncotarget 42. Rezk SA, Nathwani BN, Zhao X, Weiss LM (2013) Follicular dendritic cells: origin, function, and different disease-associated patterns. Hum Pathol 44:937–950 43. Kawabata H, Takai K, Kojima M, Nakamura N, Aoki S, Nakamura S, Kinoshita T, Masaki Y (2013) Castleman-Kojima disease (TAFRO syndrome): a novel systemic inflammatory disease characterized by a constellation of symptoms, namely, thrombocytopenia, ascites (anasarca), microcytic anemia, myelofibrosis, renal

488

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

Virchows Arch (2017) 471:467–489 dysfunction, and organomegaly: a status report and summary of Fukushima (6 June, 2012) and Nagoya meetings (22 September, 2012). J Clin Exp Hematop 53:57–61 Zinzani PL, Corradini P, Gianni AM, Federico M, Santoro A, Vitolo U, Barosi G, Tura S (2015) Brentuximab vedotin in CD30-positive lymphomas: a SIE, SIES, and GITMO position paper. Clin Lymphoma Myeloma Leuk 15:507–513 Emile JF, Diamond EL, Helias-Rodzewicz Z, Cohen-Aubart F, Charlotte F, Hyman DM, Kim E, Rampal R, Patel M, Ganzel C, Aumann S, Faucher G, Le Gall C, Leroy K, Colombat M, Kahn JE, Trad S, Nizard P, Donadieu J, Taly V, Amoura Z, Abdel-Wahab O, Haroche J (2014) Recurrent RAS and PIK3CA mutations in Erdheim-Chester disease. Blood 124:3016–3019 Diamond EL, Durham BH, Haroche J, Yao Z, Ma J, Parikh SA, Wang Z, Choi J, Kim E, Cohen-Aubart F, Lee SC, Gao Y, Micol JB, Campbell P, Walsh MP, Sylvester B, Dolgalev I, Aminova O, Heguy A, Zappile P, Nakitandwe J, Ganzel C, Dalton JD, Ellison DW, Estrada-Veras J, Lacouture M, Gahl WA, Stephens PJ, Miller VA, Ross JS, Ali SM, Briggs SR, Fasan O, Block J, Heritier S, Donadieu J, Solit DB, Hyman DM, Baselga J, Janku F, Taylor BS, Park CY, Amoura Z, Dogan A, Emile JF, Rosen N, Gruber TA, Abdel-Wahab O (2016) Diverse and targetable kinase alterations drive histiocytic neoplasms. Cancer Discov 6:154–165 Estrada-Veras JI, O’Brien KJ, Boyd LC, Dave RH, Durham BH, Xi L, Malayeri AA, Chen MJ, Gardner PJ, Alvarado Enriquez JR, Shah N, Abdel-Wahab O, Gochuico BR, Raffeld M, Jaffe ES, Gahl WA (2017) The clinical spectrum of Erdheim-Chester disease: an observational cohort study. Blood Adv 1:357–366 Mills JA, Gonzalez RG, Jaffe R (2008) Case records of the Massachusetts General Hospital. Case 25-2008. A 43-year-old man with fatigue and lesions in the pituitary and cerebellum. N Engl J Med 359:736–747 Paulli M, Bergamaschi G, Tonon L, Viglio A, Rosso R, Facchetti F, Geerts ML, Magrini U, Cazzola M (1995) Evidence for a polyclonal nature of the cell infiltrate in sinus histiocytosis with massive lymphadenopathy (Rosai-Dorfman disease). Br J Haematol 91: 415–418 Shanmugam V, Margolskee E, Kluk M, Giorgadze T, Orazi A (2016) Rosai-Dorfman disease harboring an activating KRAS K117N missense mutation. Head Neck Pathol 10:394–399 Morgan NV, Morris MR, Cangul H, Gleeson D, StraatmanIwanowska A, Davies N, Keenan S, Pasha S, Rahman F, Gentle D, Vreeswijk MP, Devilee P, Knowles MA, Ceylaner S, Trembath RC, Dalence C, Kismet E, Koseoglu V, Rossbach HC, Gissen P, Tannahill D, Maher ER (2010) Mutations in SLC29A3, encoding an equilibrative nucleoside transporter ENT3, cause a familial histiocytosis syndrome (Faisalabad histiocytosis) and familial Rosai-Dorfman disease. PLoS Genet 6:e1000833 Maric I, Pittaluga S, Dale JK, Niemela JE, Delsol G, Diment J, Rosai J, Raffeld M, Puck JM, Straus SE, Jaffe ES (2005) Histologic features of sinus histiocytosis with massive lymphadenopathy in patients with autoimmune lymphoproliferative syndrome. Am J Surg Pathol 29:903–911 Chan JK, Lamant L, Algar E, Delsol G, Tsang WY, Lee KC, Tiedemann K, Chow CW (2008) ALK+ histiocytosis: a novel type of systemic histiocytic proliferative disorder of early infancy. Blood 112:2965–2968 Facchetti F, Cigognetti M, Fisogni S, Rossi G, Lonardi S, Vermi W (2016) Neoplasms derived from plasmacytoid dendritic cells. Mod Pathol 29:98–111 Sangle NA, Schmidt RL, Patel JL, Medeiros LJ, Agarwal AM, Perkins SL, Salama ME (2014) Optimized immunohistochemical panel to differentiate myeloid sarcoma from blastic plasmacytoid dendritic cell neoplasm. Mod Pathol 27:1137–1143 Cronin DM, George TI, Reichard KK, Sundram UN (2012) Immunophenotypic analysis of myeloperoxidase-negative

leukemia cutis and blastic plasmacytoid dendritic cell neoplasm. Am J Clin Pathol 137:367–376 57. Julia F, Dalle S, Duru G, Balme B, Vergier B, Ortonne N, VignonPennamen MD, Costes-Martineau V, Lamant L, Dalac S, Delattre C, Dechelotte P, Courville P, Carlotti A, De Muret A, Fraitag S, Levy A, Mitchell A, Petrella T (2014) Blastic plasmacytoid dendritic cell neoplasms: clinico-immunohistochemical correlations in a series of 91 patients. Am J Surg Pathol 38:673–680 58. Ceribelli M, Hou ZE, Kelly PN, Huang DW, Wright G, Ganapathi K, Evbuomwan MO, Pittaluga S, Shaffer AL, Marcucci G, Forman SJ, Xiao W, Guha R, Zhang X, Ferrer M, Chaperot L, Plumas J, Jaffe ES, Thomas CJ, Reizis B, Staudt LM (2016) A druggable TCF4- and BRD4-dependent transcriptional network sustains malignancy in blastic plasmacytoid dendritic cell neoplasm. Cancer Cell 30:764–778 59. Jardin F, Ruminy P, Parmentier F, Troussard X, Vaida I, Stamatoullas A, Lepretre S, Penther D, Duval AB, Picquenot JM, Courville P, Capiod JC, Tilly H, Bastard C, Marolleau JP (2011) TET2 and TP53 mutations are frequently observed in blastic plasmacytoid dendritic cell neoplasm. Br J Haematol 153:413–416 60. Alayed K, Patel KP, Konoplev S, Singh RR, Routbort MJ, Reddy N, Pemmaraju N, Zhang L, Shaikh AA, Aladily TN, Jain N, Luthra R, Medeiros LJ, Khoury JD (2013) TET2 mutations, myelodysplastic features, and a distinct immunoprofile characterize blastic plasmacytoid dendritic cell neoplasm in the bone marrow. Am J Hematol 88:1055–1061 61. Menezes J, Acquadro F, Wiseman M, Gomez-Lopez G, Salgado RN, Talavera-Casanas JG, Buno I, Cervera JV, Montes-Moreno S, Hernandez-Rivas JM, Ayala R, Calasanz MJ, Larrayoz MJ, Brichs LF, Gonzalez-Vicent M, Pisano DG, Piris MA, Alvarez S, Cigudosa JC (2014) Exome sequencing reveals novel and recurrent mutations with clinical impact in blastic plasmacytoid dendritic cell neoplasm. Leukemia 28:823–829 62. Stenzinger A, Endris V, Pfarr N, Andrulis M, Johrens K, Klauschen F, Siebolts U, Wolf T, Koch PS, Schulz M, Hartschuh W, Goerdt S, Lennerz JK, Wickenhauser C, Klapper W, Anagnostopoulos I, Weichert W (2014) Targeted ultra-deep sequencing reveals recurrent and mutually exclusive mutations of cancer genes in blastic plasmacytoid dendritic cell neoplasm. Oncotarget 5:6404–6413 63. Varela I, Tarpey P, Raine K, Huang D, Ong CK, Stephens P, Davies H, Jones D, Lin ML, Teague J, Bignell G, Butler A, Cho J, Dalgliesh GL, Galappaththige D, Greenman C, Hardy C, Jia M, Latimer C, Lau KW, Marshall J, McLaren S, Menzies A, Mudie L, Stebbings L, Largaespada DA, Wessels LF, Richard S, Kahnoski RJ, Anema J, Tuveson DA, Perez-Mancera PA, Mustonen V, Fischer A, Adams DJ, Rust A, Chan-on W, Subimerb C, Dykema K, Furge K, Campbell PJ, Teh BT, Stratton MR, Futreal PA (2011) Exome sequencing identifies frequent mutation of the SWI/SNF complex gene PBRM1 in renal carcinoma. Nature 469:539–542 64. Sapienza MR, Fuligni F, Agostinelli C, Tripodo C, Righi S, Laginestra MA, Pileri A Jr, Mancini M, Rossi M, Ricci F, Gazzola A, Melle F, Mannu C, Ulbar F, Arpinati M, Paulli M, Maeda T, Gibellini D, Pagano L, Pimpinelli N, Santucci M, Cerroni L, Croce CM, Facchetti F, Piccaluga PP, Pileri SA (2014) Molecular profiling of blastic plasmacytoid dendritic cell neoplasm reveals a unique pattern and suggests selective sensitivity to NF-kB pathway inhibition. Leukemia 28:1606–1616 65. Dalle S, Beylot-Barry M, Bagot M, Lipsker D, Machet L, Joly P, Dompmartin A, d'Incan M, Maubec E, Grange F, Dereure O, Prey S, Barete S, Wetterwald M, Fraitag S, Petrella T (2010) Blastic plasmacytoid dendritic cell neoplasm: is transplantation the treatment of choice? Br J Dermatol 162:74–79 66. Pagano L, Valentini CG, Pulsoni A, Fisogni S, Carluccio P, Mannelli F, Lunghi M, Pica G, Onida F, Cattaneo C, Piccaluga PP, Di Bona E, Todisco E, Musto P, Spadea A, D'Arco A, Pileri S, Leone G, Amadori S, Facchetti F (2013) Blastic plasmacytoid

Virchows Arch (2017) 471:467–489 dendritic cell neoplasm with leukemic presentation: an Italian multicenter study. Haematologica 98:239–246 67. Montero J, Stephansky J, Cai T, Griffin GK, Cabal-Hierro L, Togami K, Hogdal LJ, Galinsky I, Morgan EA, Aster JC, Davids MS, LeBoeuf NR, Stone RM, Konopleva M, Pemmaraju N, Letai A, Lane AA (2016) Blastic plasmacytoid dendritic cell neoplasm is dependent on BCL2 and sensitive to venetoclax. Cancer Discov 7: 156–164 68. Geissmann F, Lepelletier Y, Fraitag S, Valladeau J, Bodemer C, Debre M, Leborgne M, Saeland S, Brousse N (2001) Differentiation of Langerhans cells in Langerhans cell histiocytosis. Blood 97:1241–1248 69. Redd L, Schmelz M, Burack WR, Cook JR, Day AW, Rimsza LM (2016) Langerhans cell histiocytosis shows distinct cytoplasmic expression of major histocompatibility class II antigens. J Hamatopathol:107–112 70. Paulli M, Locatelli F, Kindl S, Boveri E, Facchetti F, Porta F, Rosso R, Nespoli L, Magrini U (1992) Sinus histiocytosis with massive lymphoadenopathy (Rosai-Dorfman disease). Clinico-pathological analysis of a paediatric case. Eur J Pediatr 151:672–675 71. Capper D, Preusser M, Habel A, Sahm F, Ackermann U, Schindler G, Pusch S, Mechtersheimer G, Zentgraf H, von Deimling A (2011) Assessment of BRAF V600E mutation status by immunohistochemistry with a mutation-specific monoclonal antibody. Acta Neuropathol 122:11–19 72. Roden AC, Hu X, Kip S, Parrilla Castellar ER, Rumilla KM, Vrana JA, Vassallo R, Ryu JH, Yi ES (2014) BRAF V600E expression in Langerhans cell histiocytosis: clinical and immunohistochemical study on 25 pulmonary and 54 extrapulmonary cases. Am J Surg Pathol 38:548–551 73. Alayed K, Medeiros LJ, Patel KP, Zuo Z, Li S, Verma S, Galbincea J, Cason RC, Luthra R, Yin CC (2016) BRAF and MAP2K1 mutations in Langerhans cell histiocytosis: a study of 50 cases. Hum Pathol 52:61–67 74. Ritterhouse LL, Barletta JA (2015) BRAF V600E mutationspecific antibody: a review. Semin Diagn Pathol 32:400–408 75. Neumann MP, Frizzera G (1986) The coexistence of Langerhans’ cell granulomatosis and malignant lymphoma may take different forms: report of seven cases with a review of the literature. Hum Pathol 17:1060–1065 76. Egeler RM, Neglia JP, Puccetti DM, Brennan CA, Nesbit ME (1993) Association of Langerhans cell histiocytosis with malignant neoplasms. Cancer 71:865–873 77. Vasef MA, Zaatari GS, Chan WC, Sun NC, Weiss LM, Brynes RK (1995) Dendritic cell tumors associated with low-grade B-cell malignancies. Report of three cases. Am J Clin Pathol 104:696–701 78. Trebo MM, Attarbaschi A, Mann G, Minkov M, Kornmuller R, Gadner H (2005) Histiocytosis following T-acute lymphoblastic leukemia: a BFM study. Leuk Lymphoma 46:1735–1741

489 79.

80.

81.

82.

83.

84.

85.

86.

87. 88.

89.

90.

Christie LJ, Evans AT, Bray SE, Smith ME, Kernohan NM, Levison DA, Goodlad JR (2006) Lesions resembling Langerhans cell histiocytosis in association with other lymphoproliferative disorders: a reactive or neoplastic phenomenon? Hum Pathol 37:32– 39 Shao H, Xi L, Raffeld M, Feldman AL, Ketterling RP, Knudson R, Rodriguez-Canales J, Hanson J, Pittaluga S, Jaffe ES (2011) Clonally related histiocytic/dendritic cell sarcoma and chronic lymphocytic leukemia/small lymphocytic lymphoma: a study of seven cases. Mod Pathol 24:1421–1432 Feldman AL (2013) Clonal relationships between malignant lymphomas and histiocytic/dendritic cell tumors. Surg Pathol Clin 6: 619–629 Ambrosio MR, De Falco G, Rocca BJ, Barone A, Amato T, Bellan C, Lazzi S, Leoncini L (2015) Langerhans cell sarcoma following marginal zone lymphoma: expanding the knowledge on mature B cell plasticity. Virchows Arch 467:471–480 Benharroch D, Guterman G, Levy I, Shaco-Levy R (2010) High content of Langerhans cells in malignant lymphoma—incidence and significance. Virchows Arch 457:63–67 Chen W, Jaffe R, Zhang L, Hill C, Block AM, Sait S, Song B, Liu Y, Cai D (2013) Langerhans cell sarcoma arising from chronic lymphocytic lymphoma/small lymphocytic leukemia: lineage analysis and BRAF V600E mutation study. N Am J Med Sci 5:386–391 Alten J, Klapper W, Leuschner I, Eckert C, Beier R, Vallo E, Krause M, Claviez A, Vieth S, Bleckmann K, Moricke A, Schrappe M, Cario G (2015) Secondary histiocytic sarcoma may cause apparent persistence or recurrence of minimal residual disease in childhood acute lymphoblastic leukemia. Pediatr Blood Cancer 62:1656–1660 Fernandez-Pol S, Bangs CD, Cherry A, Arber DA, Gratzinger D (2016) Two cases of histiocytic sarcoma with BCL2 translocations and occult or subsequent follicular lymphoma. Hum Pathol 55:39– 43 Xie H, Ye M, Feng R, Graf T (2004) Stepwise reprogramming of B cells into macrophages. Cell 117:663–676 Ansari J, Naqash AR, Munker R, El-Osta H, Master S, Cotelingam JD, Griffiths E, Greer AH, Yin H, Peddi P, Shackelford RE (2016) Histiocytic sarcoma as a secondary malignancy: pathobiology, diagnosis, and treatment. Eur J Haematol 97:9–16 Michonneau D, Kaltenbach S, Derrieux C, Trinquand A, Brouzes C, Gibault L, North MO, Delarue R, Varet B, Emile JF, Brousse N, Hermine O (2014) BRAF (V600E) mutation in a histiocytic sarcoma arising from hairy cell leukemia. J Clin Oncol 32:e117–e121 Huang W, Qiu T, Zeng L, Zheng B, Ying J, Feng X (2016) High frequency of clonal IG and T-cell receptor gene rearrangements in histiocytic and dendritic cell neoplasms. Oncotarget 7:78355– 78362