Blastic plasmacytoid dendritic cell neoplasm: is it time to redefine the standard of care?

Keywords::

• acute lymphoid leukemia
• acute myeloid leukemia
• blastic plasmacytoid dendritic cell neoplasm
• chemotherapy

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare neoplasm currently classified among acute myeloid leukemias (AMLs) in the last edition of the WHO classification [1]. BPDCN is thought to derive from plasmacytoid dendritic cells (pDCs, also known as professional type 1 interferon producing cells or plasmacytoid monocytes), a Th2-type dendritic cell precursor subtype [2,3]. In particular, pDCs are professional effector cells capable of producing IFN-α and differentiating into dendritic cells. Physiologically, pDCs are scant in normal conditions, circulate in the blood and enter the lymph nodes and mucosal sites in response to immune stimulation [2].

In terms of its morphology, BPDCN is typically characterized by a diffuse, monomorphous infiltrate of medium-sized blastic cells with irregular nuclei, fine chromatin and one-to-several small nucleoli [1]. Despite the clinical aggressiveness, the mitotic figures, although variable in number, are rarely abundant; angioinvasion and coagulative necrosis are absent [1]. In cutaneous infiltrates, tumor cells mostly interest the dermis, sparing the epidermis, but eventually extend to subcutaneous fat. Lymph nodes are diffusely effaced in both the interfollicular areas and medulla, with the typical pattern of leukemic infiltration. Bone marrow biopsy may show either a mild interstitial infiltrate only detectable by immunophenotyping, or a massive, acute, leukemia-like infiltration; quite often, the residual hematopoiesis may present with dysplastic features, especially in megakaryocytes [4].

At immunophenotyping, the tumor cells typically express CD4, CD43, CD45RA and CD56, as well as the pDC-associated antigens CD123 (IL-3α chain receptor), BDCA-2/CD303, TCL1, the cutaneous lymphocyte-associated antigen and the IFN-α-dependent molecule MxA [1,4,5]. Rarely, CD56 can be negative, which does not exclude the diagnosis if CD4, CD123 and TCL1 are present [1]. CD68, which is physiologically found on normal pDC, is expressed in approximately half of cases, presenting as small cytoplasmic dots [1,4,5]. Among lymphoid and myeloid-associated antigens, CD4, CD7, CD22 and CD33 are more often positive, whereas CD2, CD36 and CD38 are rarely expressed. CD3, CD5, CD13, CD16, CD19, CD20, CD79a, LAT, lysozyme and MPO are mostly negative [1]. Granzyme B, which is usually found in normal pDCs, has been demonstrated at both RNA and protein levels in blood and marrow cells; but it is mostly negative on tissue sections, similarly to other cytotoxic molecules such as perforin and TIA1. Among early precursor markers, TdT is expressed in approximately 30% of cases, whereas CD34 and CD117 are negative. Accordingly, except for CD56 and TdT, the immunophenotype of BPDCN is largely consistent with that of pDCs that accumulate in reactive lymph nodes and tonsils [1]. As other hematologic malignancies (including AML, extranodal NK/T-cell lymphoma, nasal-type and other peripheral T-cell lymphomas) may express CD56 with or without CD4, an extensive immunohistochemical and/or genetic analysis is mandatory before a definitive diagnosis of BPDCN can be made [1].

The molecular genetics of BPDCN are poorly known. Specifically, metaphase cytogenetics showed that two-thirds of patients with BPDCN have an abnormal pattern, with common complex karyotypes. Conversely, specific chromosomal aberrations are lacking, while a few major recurrent chromosomal abnormalities have been recognized, including 5q21 or 5q34 (72%), 12p13 (64%), 13q13–21 (64%), 6q23 (50%), 15q (43%) and loss of chromosome 9 (28%) [1,6]. In addition, comparative genomic hybridization analysis indicated the recurrent involvement of regions on chromosome 4 (4q34), 9(9p13–p11 and 9q12–q34) and 13 (13q12–q31), with common alteration of the loci where p18, p16, p27 and RB are encoded [1,7]. However, so far, neither has the deregulation of candidate genes been clearly demonstrated, nor have functional experiments been carried on. On the other hand, more recently, the occurrence of TET2 and TP53 somatic mutations was documented, although the functional consequences of such abnormalities has not yet be determined [8]. Finally, gene-expression profile studies provided evidence that BPDCN is distinct from other myeloid cutaneous diseases, including acute myelomonocytic leukemia [9].

On the clinical ground, although classified among AMLs, BPDCN presents as leukemic disease only in a minority of cases. On the contrary, it more frequently involves the skin and, eventually, in more advanced stages, lymph nodes and other structures [1]. Of note, the disease is very aggressive, being characterized by scarce response to conventional chemotherapies and rare long-term disease-free and overall survivals [4,10]. To date, also based on its classification, the most commonly used approaches are AML-like chemotherapy regimens and, when possible, autologous or allogeneic stem cell transplantation [10–14]. The latter, in particular, appeared to be very relevant for inducing long-term complete remissions (CRs) and should then also be considered in elderly patients, which actually represent the majority of cases [13]. On the other hand, the choice of the induction/consolidation regimens probably warrants some further considerations. In fact, although no clinical trials have been reported on BPDCN, the routine usage of AML-like chemotherapy may not always be the optimal strategy. On the contrary, some evidence would rather support the usage of acute lymphoblastic leukemia (ALL)-like protocols. In this regard, recent reports on national series indicated the possible superiority of these schemes. First, Tsagarakis et al. described a series of 19 BPDCNs, treated with either AML-like (n = 6) or ALL-like schemes (n = 9) [15]. Noteworthy, a relative advantage of ALL-type regimens was documented in this series. Specifically, all nine patients who received ALL-type induction therapy achieved CR (vs three out of six patients with AML type), and relapse appeared to occur less frequently in patients receiving this type of therapy. Interestingly, as the majority of long-term survivors also received an allogeneic stem cell transplantation, it is likely that the use of regimes inducing a higher rate of CR could be useful for bridging the patients to that procedure. However, the number of patients receiving AML-type therapy and achieving CR was too small to draw clear conclusions [15]. Similarly, Pagano et al., for the Gruppo Italiano Malattie Ematologiche Maligne dell’Adulto (GIMEMA) group, made an analogue observation in a large series recently collected in Italy [16]. Moreover, a recent analysis focusing on pediatric cases indicated a quite brilliant efficacy of ALL-like schemes in children affected by BPDCN [17].

Unfortunately, as mentioned previously, there are no randomized clinical trials supporting the usage of these specific regimens. However, some biological issues may offer the rational basis for this approach. In fact, although pDCs are commonly considered within the myeloid compartment, their derivation is quite complex (see Figure 1 in the paper by O’Garra et al.) [18,19]. Specifically, pDCs can derive from both myeloid and lymphoid progenitors, the latter option possibly being the favorite one during physiological development. In addition, BPDCNs usually lack most myeloid antigens, while expressing lymphoid molecules such as CD4, CD22 and CD123 [1,20]. Therefore, it is conceivable that BPDCN cells may retain some biological features of lymphoid elements, requiring ad-hoc therapies. In this regard, more refined molecular analyses, including massive parallel sequencing, are definitely warranted to gain further insights into the pathobiology of this disease. Indeed, such approaches may also lead to the identification of novel potential therapeutic targets, opening novel scenarios in the management of the disease. On the other hand, dedicated clinical trials are necessary to offer a solid, evidence-based therapeutic indication. In this view, certainly, due to the rarity of the disease, the creation of national (continental) registries would definitely be warranted.

Financial & competing interests disclosure

This study was financially supported by BolognAIL, AIRC (IG4987; 5xMille 10007; IG10519), RFO (SA Pileri and PP Piccaluga), Fondazione Cassa di Risparmio in Bologna, Fondazione della Banca del Monte e Ravenna and Progetto Strategico di Ateneo 2006 (SA Pileri and PP Piccaluga). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.