Institution: University of Alabama at Birmingham
Additional authors:Reddy, Vishnu VB; Wakefield, Dara; Peker, Deniz
Session: Extramedullary manifestations of myeloid neoplasms
HISTORY
28 year old male presented with a 3 month history of recurrent headaches, facial pain, and sinusitis, in July, 2012. At UAB, the radiologist’s interpretation of outside imaging studies identified soft tissue opacification concerning for a sinonasal mass which appear to arise within the right mid nasal cavity and extends into the proximal right maxillary sinus opacifying the antrum and potentially into the right sphenopalantine fossa. Not appearing aggressive enough to represent sinonasal undifferentiated carcinoma or angiofibroma, the differential diagnosis included a squamous cell lesion, with a papilloma being less likely. The patient underwent an operative procedure including right nasal endoscopy with maxillary sinusotomy and mucous membrane removal, right partial ethmoidectomy, septoplasty and frameless stereotactic computer assisted surgical navigation. A frozen section was performed and a preliminary diagnosis of small round blue cell tumor was reported. Based on histologic, flow cytometry, immunohistochemical, and in-situ hybridization studies a final diagnosis of blastic plasmacytoid dendritic cell neoplasm (BPDCN) was rendered. Subsequent flow cytometry performed on cerebrospinal fluid (CSF) showed involvement by BPDCN. A staging bone marrow biopsy was negative. A 1 mm nodule on the right front parietal scalp, thought to represent skin involvement, was biopsied after the initial round of chemotherapy and was diagnosed as a compound nevus with a congenital pattern. There was no other suspicious skin lesions identified. Status post intrathecal chemotherapy, consisting of cytarabine and hydrocortisone for CSF involvement, induction chemotherapy with a 7 + 3 regimen consisting of cytarabine and idarubicin, and consolidation treatment with HIDAC #1, the patient was considered in complete remission and proceeded to allogeneic matched unrelated donor peripheral blood stem cell transplant. Six months after diagnosis the patient remains free of disease.
DETAILS
The biopsy was performed on the right ethmoid sinus and fixed in 10% neutral buffered formalin for at least 8 hours. 3-5 micron sections were stained with hematoxylin and eosin (H&E) stain.
The H&E sections illustrated an excisional biopsy of sinus mucosa and underlying submucosal tissue involved by an extensive neoplastic infiltrate consisting of medium sized cells with high nuclear to cytoplasmic ratio, blastic chromatin pattern, inconspicuous nucleoli and scant cytoplasm (Figures 1, 2a, and 2b). Significant mitotic activity, angiocentrism, angioinvasion, and necrosis were not identified. Scattered background inflammatory cells were present including plasma cells and small lymphocytes. Focally, a grenz-like zone can be seen between neoplastic cells and the overlying surface sinus mucosa, similar to cases with skin involvement where a grenz zone is present between neoplastic cell and the epidermis (Figure 3a and 3b). The staging bone marrow showed normal trilineal hematopoiesis with no morphologic or phenotypic evidence of BPDCN (Figure 4). Subsequent to initial chemotherapy, a biopsy of the scalp lesions showed a compound nevus with a congenital pattern (Figure 5). There was no morphologic evidence of BPDCN, which was confirmed by negative immunoperoxidase stain for CD123 (Figure 6).IMMUNOHISTOCHEMISTRY AND FLOW CYTOMETRY
Flow cytometry analysis of the right ethmoid mass shows a population of neoplastic cells which are CD4 dim, CD56 high, CD45lo, CD117 dim, TdT+, HLA-DR + (Figure 7a-f). This population did not express T-cell markers (CD5, CD7, CD8, surface or cytoplasmic CD3), B cell markers (CD19, CD20, CD23), myeloid markers (CD13, CD33, CD34, CD15, cytoplasmic MPO), monocytic markers (CD14, CD64), natural killer cell markers (CD16, CD57) or plasma cell markers (CD138, CD38).
By immunohistochemistry, performed on the ethmoid sinus lesion, CD123 (Figure 8a and 8b) and TdT (Figure 9a) is diffusely positive in tumor cells and CD4 is weakly positive. A Ki-67 shows a high proliferative index of 70-80% (Figure 9b).Flow cytometry analysis of the cerebrospinal fluid shows that a proportion of the cellular events (44% of cells) consist of an aberrant CD117+, CD56+, HLA-DR+ positive population, similar to the ethmoid sinus mass (Figure 10a-10c).CYTOGENETIC FINDINGS
Normal male karyotype (46, XY; performed on bone marrow).
MOLECULAR FINDINGS
In situ hybridization study for Epstein Barr encoded RNA (EBER) is negative (Figure 11).
INTERESTING FEATURES
The current case is a very rare presentation of an uncommon hematopoietic neoplasm with an indolent initial appearance in a young patient. The radiologic studies were non-specific and biopsy revealed a blastoid neoplastic infiltration with no angioinvasion/destruction. The differential diagnosis of BPDCN can vary depending on the site involvement, however in most cases immunophenotypic studies allow for sufficient discrimination. When detected in the blood and bone marrow, diagnostic considerations may include acute myeloid leukemia, which share myelomonocytic markers and precursor lymphoblastic leukemia/lymphoma, which can express TDT, but are both otherwise distinguished from BPDCN by a comprehensive panel. Skin involvement can mimic T-cell lymphomas, excluded by absent expression of surface CD3, cutaneous NK-cell lymphoma, ruled out by absent expression of cytotoxic proteins, and myelomonocytic subtype leukemia cutis, which may not always be reliably excluded. Unique to our case are several diagnostic entities not infrequently sampled at the time of frozen section, commonly seen in the sinonasal tract, and easily misdiagnosed if flow cytometry is not requested. Several such as nasopharyngeal carcinoma, olfactory neuroblastoma, and sinonasal carcinomas usually express epithelial markers. Ewing Sarcoma/PNET will express CD99 or FLI-1. Nasal type extranodal T/NK cell leukemia can share CD56 expression with BPDCN and it is critical to order testing for EBV such as with in situ hybridization studies, which is never positive in BPDCN. Additionally nasal type T/NK cell leukemia will usually demonstrate angioinvasion and prominent necrosis. Most of the 200 cases of BPDCN described in the literature have cutaneous involvement; therefore BPDCN without cutaneous involvement is a distinctly unusual finding, often restricted to a leukemic phase in peripheral blood and/or lymph node involvement. Only a handful of cases have been reported with BPDCN limited to a non-hematopoietic site and with no skin involvement. To the best of the authors’ knowledge this report represents the first reported case presenting with isolated sinonasal involvement including contiguous CNS involvement, with no known skin, peripheral blood, bone marrow or lymph node findings on physical exam. Our patient’s right ethmoid sinus lesion, as described above matches previous histologic description of BPDCN, with the additional finding of a grenz like zone between submucosal tumor cells and the surface mucosa.
This case represents a good example of the diagnostic challenges surrounding a relatively newly described and still evolving entity. A high index of suspicion is often needed in order to perform the testing necessary to diagnose BPDCN, especially when the presentation is atypical. By flow cytometry or immunohistochemical staining, finding co-expression of CD4, CD56, and CD123 in the absence of other lineage specific antigens is often critical to appropriate diagnosis. In the rare difficult or inconclusive cases, where one or more of the above antigens is not detected, testing for other plasmacytoid dendritic cell antigens such as TCL1, CD2AP and BDCA1/CD303 can be performed. In some cases cytochemical stains or an EBER may be helpful in excluding other neoplasms. It is especially important to distinguish BPDCN from acute myeloid or lymphoblastic neoplasms since the former will often initially respond well to treatments for the latter, before almost always recurring. An appropriate diagnosis of BPDCN may possibly improve a patient’s survival by allowing the clinical team to plan a treatment regimen aimed toward the goal of getting the patient an allogeneic stem cell transplant.PROPOSED DIAGNOSIS
Blastic Plasmacytoid Dendritic Cell Neoplasm (BPDCN)
CONSENSUS GROUP: ADDITIONAL INFORMATION/STUDIES
Additional immunohistochemical stains performed by the consensus group:
TCL1: Negative
CONSENSUS DIAGNOSIS
Blastic plasmacytoid dendritic cell neoplasm