BPDCN: state of the art Free

A 67-year-old man with no significant past medical history (including no history of cardiopulmonary disease) and ECOG Performance Status (PS) of 1 presents with easy bruising, increasing fatigue, palpable lymphadenopathy, and rapidly growing purplish/violaceous discolored skin lesions with necrotic areas on the arms, trunk, and face. He was initially given steroids and antibiotics by his primary care provider, with transient resolution of some of the skin lesions; however, within approximately 3 weeks, the skin lesions reappeared and in greater number. Further workup revealed electrolytes, creatinine, and liver tests within normal limits; an albumin normal range at 4.1 g/dL; and pancytopenia with 15% circulating blasts. Echocardiogram demonstrated an ejection fraction of 60%. The patient was then referred to hematology. The patient underwent bone marrow biopsy, positron emission tomography (PET)/computed tomography (CT) scan, and skin biopsy. The combined results of the skin and bone marrow biopsy with flow cytometry and immunohistochemistry demonstrated malignant immature plasmacytoid dendritic cells with positivity for CD4, CD56, CD123, TCL1, CD303, and TCF4. PET/CT scan revealed several areas of adenopathy above and below the diaphragm. A lumbar puncture was performed, which was negative for malignant cells. The patient was given a diagnosis of blastic plasmacytoid dendritic cell neoplasm, or BPDCN. He was administered upfront therapy with tagraxofusp, an approved CD123-targeted agent for patients with BPDCN. In cycle 1, the patient experienced hypoalbuminemia (managed with IV albumin and furosemide administration), transient thrombocytopenia (with no bleeding) and 2 times the upper limit of normal elevations of aspartate transaminase and alanine transaminase. By cycle 2, all these had resolved. The patient achieved complete remission after 2 cycles (as documented by negative bone marrow, PET/CT negativity, resolution of skin lesions with negative biopsy, and lumbar puncture remaining CSF negative for malignant cells) and remained on tagraxofusp for a total of 4 cycles, after which time the patient was referred for consideration of an allogeneic stem cell transplantation (SCT). The patient ultimately underwent matched unrelated donor allogeneic SCT and is doing well 3.5 years later.

BPDCN is a unique hematologic malignancy that features elements of both myeloid and lymphoid malignancies and exhibits a rather aggressive clinical phenotype and historically difficult course in terms of patient management.^1-3 BPDCN is characterized by involvement of 4 primary organ compartments, including cutaneous sites, bone marrow/blood involvement, lymph node involvement, and a remarkably high incidence of central nervous system/cerebrospinal fluid (CSF) involvement.⁴ The triad of expression of the IL-3Rα surface marker (CD123+), as well as CD4+ and CD56+, remains the most well-known aspect of BPDCN diagnosis in most cases. Increasing the ability to aim for 100% specificity, the additions of TCL1, TCF4, and CD303 have greatly assisted in diagnosing this complex rare malignancy.⁵ At the molecular level, BPDCN tends to be similar to that of myeloid malignancies, such as myelodysplastic syndrome (MDS), chronic myelomonocytic leukemia (CMML), or acute myeloid leukemia (AML), as the most common mutations or variants that groups have repeatedly identified are TET2, ASXL1, RAS, and splicing factor mutations such as ZRSR2⁷ have been identified. Risk factors for organ involvement of BPDCN, still being understood by the field, appear to include male sex, UV light sun exposure, smoking, and association in older patients with TET2, RAS, and other molecular mutations.^8-10 

The exact incidence and prevalence of BPDCN worldwide remains unknown. With its true incidence in epidemiology is difficult to capture given the historical rarity of the disease and the many nomenclature changes throughout the past several decades (Table 1), it is believed that BPDCN accounts for approximately 0.44% of all hematologic malignancies, and one database study found that BPDCN accounts for 0.04 cases per 100,000.^11,12 

Most reports prior to the modern targeted-therapy era demonstrated a median overall survival of approximately 8 to 16 months. Most of these patients either presented with or ultimately progressed to acute leukemic transformation.¹ Most of these patients were treated with cytotoxic chemotherapy and regimens such as AML-based, acute lymphoblastic leukemia (ALL)-based, cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP)-lymphoma-based, multiple myeloma–based, and even local skin-directed therapies.¹³ Despite multimodality chemotherapy regimens, the prognosis for relapsed/refractory (R/R) BPDCN remains poor, and this has led to the urgent unmet medical need of searching for new targets, new approaches, and new combinatorial therapy, including those that cross the blood-brain barrier.¹ 

From the time a diagnosis of BPDCN is suspected, it is imperative to have a team communicating together, including the hematopathologist, dermatopathologist, hematologist/oncologist, stem cell transplant team, and dermatologist. One widely accepted multidisciplinary approach has been published by the North American Blastic Plasmacytoid Dendritic Cell Neoplasm Consortium (NABC).¹⁴ This outline for the diagnostic evaluation of BPDCN includes a baseline cutaneous and systemic patient evaluation with emphasis on dermatologic consultation featuring early and serial skin biopsies of target lesions, and inclusion of a detailed visual examination of the skin with medical photography for reference. The preference in the field has been to use the mSWAT skin scoring system used in mycosis fungoides. In addition, baseline bone marrow aspiration and biopsy including morphology, blasts percentage, cytogenetics, molecular mutation next-generation sequencing panel, and, importantly, inclusion of both immunohistochemistry markers and flow cytometry panel specifically for BPDCN, including CD123, CD4, CD56, CD303, TCL1, and TCF4. Baseline evaluation with a PET/CT or CT scan for assessment of lymph nodes or other sites of extramedullary involvement is highly recommended for all patients with BPDCN given the disease's high propensity toward extramedullary involvement. Finally, patients with BPDCN should have a baseline lumbar puncture to detect for CSF evaluation/CNS disease given the high incidence of CSF involvement in BPDCN even at baseline.¹⁴ 

As the vast majority of cases of BPDCN will feature cutaneous involvement at some point in the disease course, one important aspect of this particular rare disease topic is to emphasize the broader differential diagnosis for patients suspected of having a skin-involving hematologic malignancy. On a practical note, it is a good idea to keep BPDCN in the differential diagnosis for all patients undergoing a hematologic workup in the setting of abnormal skin lesions. One of the more common presentations of BPDCN remains the patient who is initially given a series of other therapies for an unknown acquired cutaneous lesion with either antibiotics or steroid-based regimens, in some cases leading to partial regression of the lesions and of course subsequent diagnostic delay, as in our patient in the opening case presentation. Despite these therapies, the persistence or rapidly growing lesion will ultimately lead to a skin biopsy, but it often may be done too far late in the disease course and perhaps in the setting of more advanced/leukemic features, and so early suspicion, early skin biopsy, and early recognition are keys in BPDCN diagnosis and management (Figure 1).

In terms of a more general differential diagnosis, one must keep in mind and distinguish among malignant, paraneoplastic, and nonmalignant etiologies involving the intersection of the skin and blood, metastatic melanoma, cutaneous T-cell lymphomas, AML with leukemia cutis/myeloid sarcoma, CMML with paraneoplastic skin manifestations, Sweet's syndrome, fungal/infections, skin complications, autoimmune skin manifestations, and drug/ allergic reactions. At a deeper level, once a plasmacytoid dendritic cell neoplasm is identified, it is imperative to distinguish among the various entities with this category: BPDCN vs mature pDC proliferation (MPDCP) vs pDC-AML, which can be done with collaboration by expert hematopathology/flow cytometry and dermatopathology teams in coordination with the clinical team.^12,14,15 

Prior to the current modern era of BPDCN targeted therapy, the most common therapy regimens for the younger/fit patient with BPDCN included (1) ALL (hyper-CVAD/CVD +/– ven, or other ALL induction regimens)-type therapies, (2) AML based (eg, 7 + 3), (3) lymphoma based (CHOP), or (4) multiple myeloma based. Many of these historical therapies are nicely summarized by Taylor et al.¹³ Notably, the vast majority of patients with BPDCN who present around the world are older, have serious comorbidities, and are often frail or unfit for intensive chemotherapy. These patients have historically been treated with various other strategies, including lower-intensity AML or lower-intensity ALL approaches.^16,17 As we move into the modern targeted era, it is an important practice point to emphasize that these therapy options still remain absolutely appropriate first-line therapies when needed, especially across the world in areas that do not have access to targeted therapies; notably, this keeps the option for CD123-targeted and BCL2-targeted therapies to be used in second-line settings and beyond, including after referral for specialty consultation once the patient is stabilized.

A pilot study of 11 male patients with BPDCN was conducted by Frankel et al to determine the efficacy of tagraxofusp (SL-401).¹⁸ The median age was 70 years. Seven of 9 evaluable patients had a major response, which included 5 complete remissions and 2 partial remissions. The major toxicity from this novel agent was capillary leak syndrome (CLS). After the success of this pilot study, Pemmaraju et al conducted a larger study of tagraxofusp on patients with BPDCN, this time of 47 patients in a clinical trial in 4 stages (stage 1 was a lead-in dose escalation; stage 2, expansion; stage 3 pivotal confirmatory; stage 4 extension cohort). The investigators demonstrated a 90% overall response rate among frontline-treated patients (72% complete response [CR]/CRC), as well as a 67% overall response rate among relapsed/refractory patients. The major toxicity, as expected from the prior study, was indeed CLS, which was found to be potentially fatal and received a black box warning from the Food and Drug Administration (FDA). This novel agent was ultimately approved by the FDA in December 2018, representing the first targeted drug approved for BPDCN and importantly the first overall approved CD123-targeted agent in the entire field of hematology/oncology.¹⁹ Subsequently, tagraxofusp was approved by the European Medicines Agency for BPDCN as a frontline treatment for adult patients in January 2021.²⁰ Our 67-year-old patient with BPDCN from the opening case was chosen for tagraxofusp therapy on the basis of his fit PS, lack of cardiopulmonary comorbidities, albumin of ≥3.2 g/dL, and central nervous system negativity, in keeping with the original clinical trial parameters from Pemmaraju et al.¹⁹ His case was in line with what was expected for tagraxofusp administration, including highlighting the transient nature of decrease in albumin and platelets and increased LFTs; the vast majority of cases, as in our patient, are generally restricted to the first cycle of therapy. In the original New England Journal of Medicine study, 45% of frontline patients, regardless of older median age, as was the case with this patient, were able to be transitioned to SCT.¹⁹ Longer-term follow-up results, with median follow-up of approximately 3 years, continued to show high rates of efficacy, particularly in frontline-treated patients, as well as no new safety signals. Importantly, in modern standard-of-care practice, patient selection remains key to selecting appropriate patients for tagraxofusp based on the required clinical aspects of pre-dosing, including adequate albumin, creatinine, and cardiopulmonary patient parameters, as well an emphasis on monitoring daily weights, creatinine, liver function tests, fluid status, and cardiopulmonary status during and after drug administration days.²¹ 

Another encouraging CD123-targeted approach for BPDCN that remains in active late-stage clinical development is that of pivekimab sunirine (PVEK). PVEK is a first-in-class antibody-drug conjugate made up of a CD123-antibody, a novel linker, and a unique cytotoxic payload indolinobenzodiazepine dimer (IGN). The IGN payload alkylates DNA and causes single-strand breaks without cross-linking. IGNs are designed to have high potency against tumor cells while demonstrating less toxicity to normal marrow progenitors than other DNA-targeting payloads.²² Clinically, Daver et al have demonstrated promising results for PVEK in patients with R/R AML.²³ For BPDCN, PVEK was investigated in the CADENZA (IMGN632-801): open-label, multicenter, phase 2 study (NCT03386513). As of the European Hematology Association 2023 presentation by Pemmaraju et al, among 30 frontline-treated patients (median age 74 years, range 48-84 years), there was an 80% objective response rate (24 of 30) with a composite CR of 73% (22 of 30). The time to first response was a median of 1.3 months (0.5 to 3.5 months), and 30% were able to be bridged to allogeneic SCT. There were no CLS cases reported in this frontline cohort. Among the total n  =  79 treated, the most common adverse events included peripheral edema in 46%, thrombocytopenia in 27%, and infusion-related reactions in 25%. Among notable adverse events were liver-related adverse events: alanine transaminase/aspartate transaminase/total bilirubin laboratory elevations were grade 3 in 3%, 3%, and 1%, respectively; there were no grade 4 or 5 events. There was a 0% frontline mortality rate in the first 30 days, and median duration of response was 12.7 months, including post-SCT durability.²⁴ This agent remains under active investigation in the BPDCN field.

The curative potential for allogeneic SCT has been solidly demonstrated for patients with BPDCN, especially if performed in CR1.^25,26 The field is still investigating the optimal types of SCT, the optimal depth of response necessary prior to SCT, and the ongoing intriguing question of benefit in a limited or select group of patients with an autologous SCT approach.^27-29 And so worldwide, the field has recognized that SCT is the only fully curative option in patients with BPDCN and should be offered for consideration to all patients during their BPDCN journey when feasible. One important clinical pearl is that a patient's ECOG PS is a dynamic measure and often can actually improve over time in some patients with various treatments, particularly in the setting of a massive tumor burden at presentation that is greatly reduced; therefore, a patient can be reconsidered for SCT later during the treatment course if they had previously been excluded due to PS or comorbidities.³⁰ 

One of the most unexpected aspects of modern-day care in BPDCN is the recognition of central nervous system (CNS) involvement in BPDCN.³¹ Most often manifested as asymptomatic CSF+ on lumbar puncture (LP) testing, it can also range to debilitating or life-threatening involvement, including meningeal spread³² and even ocular invasion of BPDCN.^33-35 Most studies have found the incidence of CNS involvement in BPDCN to range from 20% to 30%, which remarkably includes not only R/R but also frontline, mostly asymptomatic cases.³⁶ In this setting, it is now a mandatory part of BPDCN care to offer an LP with IT chemotherapy at baseline and multiple LPs for CNS prophylaxis regardless of the BPDCN treatment chosen, especially as the modern targeted therapies do not appear to cross the blood-brain barrier.³⁷ One widely used modern approach by Pemmaraju et al features 8 LPs with alternating intrathecal (IT) cytarabine (ARA-C) and IT methotrexate up to 12 LPs in higher-risk patients.⁴ And so, Central nervous system (CNS) involvement by BPDCN is so surprisingly frequent that this finding requires a worldwide paradigm-shifting change in practice to offer some form of prophylactic/therapeutic intervention to all patients with BPDCN.³⁸ 

Although BPDCN is generally regarded as a disease of older patients, a recently emerging theme is the increasing emphasis on identifying and treating younger patients with this rare hematologic malignancy.^39,40 A growing proportion of pediatric and adolescents and young adults (AYA) patients now constitute the overall BPDCN group.⁴¹ Furthermore, there are cytogenetic and molecular differences in older BPDCN patients vs their younger counterparts in the setting of myriad different personal medical histories (clonal hematopoiesis of intederminate potential, prior or concomitant hematologic malignancies), environmental exposures (smoking, radiation, ultra-violet light⁹), and incidence of molecular mutations (eg, MYB rearrangements among pediatric BPDCN^42,43). Among 3 R/R BPDCN patients treated by Sun et al with tagraxofusp, 2 of 3 experienced rapid (although transient) responses as monotherapy, and ultimately tagraxofusp attained US FDA approval, including for patients age 2 and older.⁴⁴ Among 8 patients treated by Pemmaraju et al with tagraxofusp for pediatric/AYA BPDCN (ages 2-21 years), the majority of patients of interest (88%) were female. Five patients were treated in the frontline setting and 3 in the R/R setting; 33% (3 of 8) achieved CR, and 63% (5 of 8) were bridged to SCT.⁴⁵ Cytotoxic chemotherapy programs remain widely used treatment approaches in pediatric/AYA patients. One such therapy is HCVAD-based regimens. He et al treated an AYA BPDCN patient with extensive initial presentation involving CNS and orbital involvement with HCVAD+VEN; the patient achieved CR1 and was able to undergo allogeneic SCT.⁴⁶ Abla et al treated an 11-year-old patient in Lebanon with R/R BPDCN status post SCT with HCVAD+VEN in the setting of testicular involvement. The patient achieved CR2 and was able to go on to second allogeneic SCT in the setting of no available CD123-targeted therapies.⁴⁷ 

Montero et al found that BPDCN is highly dependent on BCL2. Preclinical studies have confirmed the sensitivity of BPDCN to venetoclax. Lane and Pemmaraju treated 2 older/unfit patients with BPDCN in the R/R setting with venetoclax monotherapy, both yielding rapid yet overall transient responses.⁴⁸ Dinardo et al demonstrated that an hypomethylating agent (Decitiabine or Azacytidine)+venetoclax combination could lead to clinical responses in off-protocol patients with R/R BPDCN,⁴⁹ and Gangat et al further demonstrated that off-protocol older/ unfit patients with BPDCN even in the frontline setting can benefit from HMA+VEN (n  =  10 patients, all having some form of transient response, with n  =  2 actually proceeding to SCT and improving their PS from baseline presentation).³⁰ Togami et al performed a crucial study demonstrating that TAG resistance in BPDCN and AML is often due to acquired reductions in the diphthamide synthesis pathway, which is uniquely overcome by HMA (azacitidine) therapy, hence leading to clinical rationale for HMA+TAG combination therapy approaches.⁵⁰ The triplet approach of TAG/AZA/VEN, already investigated in AML by Lane et al, is now being investigated in for patients with both R/R and frontline BPDCN.⁵¹ Further efforts to combine BCL2-targeted therapy with TAG and cytotoxic chemotherapy (HCVAD)^52,53 have led to the ongoing clinical trial of total therapy with TAG/VEN/HCVAD currently actively enrolling for both frontline and R/R BPDCN. (See Table 2 for a selection of active clinical trials in BPDCN.)

In the growing BPDCN field, several national and international collaborations have been forming. In groups around the world, from adult and pediatric groups in the United States, Canada, and Mexico (North America), to consensus groups in Italy dedicated to BPDCN, to most recently an international registry and collaboration led by Gevorg Tamamyan and the Immune Oncology Research Institute team from Armenia, we are starting to see an infusion of new ideas and collaborations in the BPDCN field. As these groups are now beginning to publish their findings, the BPDCN field has received an exponential increase of clinical awareness and increased research efforts across institutions and countries. Table 3 highlights a selection of some of the more recent consortia.^{13,14,40,54-56} 

This article highlights the past, present, and future directions for BPDCN therapy. At present, there are several standard-of-care options for patients with BPDCN depending on ECOG PS and availability of clinical trials (always preferred as a first choice in this ultra rare disease) and/or targeted agents. In the younger/fit patient with BPDCN, worldwide standards of care now include ALL-based regimens, myeloid-based regimens, and CD123-targeted agents on or off clinical trials. The addition of VEN when available to cytotoxic chemotherapy–based regimens has been demonstrated. In older/unfit BPDCN patients, an HMA+VEN approach or lower-intensity ALL-based options are considered standard of care, and again, CD123-targeted agents can be considered if available. CNS prophylaxis should be offered to all patients with BPDCN. All patients who can be considered for SCT should be offered it in CR1. Future directions will include further evaluating and understanding triplet combinations (CD123/HMA/VEN and CD123/chemotherapy/VEN combinations in older/unfit and younger/fit patients, respectively); further defining the role of the number and frequency of lumbar punctures with IT chemotherapy pre and post SCT; further defining the role and type of SCT (allo- vs auto-, types of donors, etc); and identifying new targets and pathways beyond CD123 and BCL2, all of which will be the purview of international consortia and collaborations over the next decade.

This study was supported in part by the MD Anderson Cancer Center Support Grant (CCSG) CA016672 and the MD Anderson Cancer Center Leukemia SPORE CA100632.

Naveen Pemmaraju: consultancy/scientific advisory board/ speaking: Pacylex Pharmaceuticals, Astellas Pharma US, Aplastic Anemia and MDS International Foundation, CareDx, ImmunoGen, Inc, Bristol Myers Squibb Co., Cimeio Therapeutics AG, EUSA Pharma, Menarini Group, Blueprint Medicines, CTI BioPharma, ClearView Healthcare Partners, Novartis Pharmaceuticals, Neopharm, Celgene Corporation, AbbVie Pharmaceuticals, PharmaEssentia, Curio Science, DAVA Oncology, Imedex, Intellisphere, CancerNet, Harborside Press, Karyopharm, Aptitude Health, Medscape, Magdalen Medical Publishing, MorphoSys, OncLive, Patient Power, Physicians' Education Resource (PER), PeerView Institute for Medical Education; research funding (grant): United States Department of Defense, National Institutes of Health/ National Cancer Institute (NIH/NCI); membership on board of directors/management: Dan's House of Hope; leadership: ASH Committee on Communications, ASCO Cancer.Net Editorial Board; licenses: Karger Publishers; uncompensated: HemOnc Times/Oncology Times.

Naveen Pemmaraju: As BPDCN is a rare disease, please note there is discussion of potentially off-label/novel uses of therapies in this manuscript, supported by relevant litetaure.