Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare and aggressive hematological malignancy that primarily involves the skin, bone marrow (BM), and lymph nodes. With an estimated annual incidence of less than 1 per million, BPDCN predominantly affects elderly males and is associated with a dismal prognosis, with median survival ranging from 12 to 16 months despite intensive treatment regimens (1). The classification of BPDCN has evolved significantly over the years, and it is now categorized under dendritic cell and histiocytic tumors in the most recent 5th World Health Organization (WHO) Classification of Hematolymphoid Tumors (WHO-HEMA5). In WHO-HEMA5, the classification of histiocytic/dendritic cell neoplasms is placed after myeloid tumors because they originate from a common myeloid progenitor cell, forming the monocyte/histiocyte/dendritic cell lineage. This update reflects ongoing efforts to better understand the pathogenesis and cellular origins of this enigmatic neoplasm (2). Secondary BPDCN, which arises in the context of prior or concomitant hematologic malignancies (PCHM), has been reported in 10%–20% of cases (3). However, the relationship between clonal mutations in these two malignancies remains poorly understood. Recent studies have identified recurrent mutations in genes such as TET2, ASXL1, ZRSR2, and RAS, which are frequently observed in BPDCN (4, 5). Here, we report a case of BPDCN secondary to acute myeloid leukemia (AML) and provide a comprehensive review of cases in which the clonal relationship between BPDCN and associated hematologic neoplasms has been clearly established. This aims to deepen our understanding of this rare phenomenon and discuss the potential clonal relationships between different tumors.

A 75-year-old male patient initially presented with fatigue and fever. Laboratory investigations revealed a significantly elevated white blood cell count of 39 × 10^9/L, a hemoglobin level of 102 g/L, and a platelet count of 62 × 10^9/L. Peripheral blood smear analysis revealed 86% blasts, suggestive of a hematologic malignancy. BM examination showed marked hypercellularity, with substantial suppression of both the granulocytic and erythroid lineages (Figure 1A). Blasts accounted for 97% of the marrow cellularity. Flow cytometry of the BM revealed that 83% of the cells expressed CD13, CD33, HLA-DR, CD117, CD38, CD6, CD7, and CD123, with partial expression of CD36 and CD371. These findings were consistent with the presence of malignant myeloid blasts (Figure 1C). Based on these investigations, a diagnosis of acute myeloid leukemia (AML) was established. Next-generation sequencing (NGS) revealed mutations in DNMT3A (41.06%), TET2 (43.32%), NRAS (1.07%), and TP53 (1.82%). Cytogenetic analysis showed a normal karyotype. The patient, classified as fit for intensive chemotherapy, was treated with the “7 + 3” regimen, which consists of 7 days of cytarabine and 3 days of idarubicin. Following this regimen, the patient achieved complete remission. Post-treatment flow cytometry analysis revealed no evidence of minimal residual disease. The patient and his family declined allogeneic hematopoietic stem cell transplantation. The patient subsequently underwent two additional cycles of intensive chemotherapy, followed by three courses of maintenance therapy with azacitidine and venetoclax.

Nine months after the initial diagnosis, the patient presented with new-onset vision impairment and unsteadiness in gait, which had developed over the past week. Upon admission, laboratory findings showed a white blood cell count of 1.29 × 10^9/L, hemoglobin of 115 g/L, and platelet count of 64 × 10^9/L. Peripheral blood smear again revealed 86% blasts, raising concern for AML relapse. The bone marrow morphology report revealed that unclassifiable cells accounted for 18% of nucleated cells (Figure 1B).​​ ​These cells exhibited medium size, with a round to irregular shape.​​ ​Their cytoplasm ranged from scanty to moderate in amount and displayed a characteristic basophilic (grayish-blue) hue, occasionally featuring a perinuclear halo.​​ ​The nuclear chromatin appeared coarse, with inconspicuous nucleoli.​​ ​Notably, some cells demonstrated pseudopodia and tailing projections.​​ ​Collectively, these morphological features raise strong suspicion of BPDCN.​ Immunophenotyping demonstrated the expression of markers consistent with BPDCN, including HLA-DR, CD4, CD7, CD38, CD56, CD117, CD123, and CD304, with partial positivity for CD33 (Figure 1D). These findings were suggestive of BPDCN rather than AML relapses. Notably, the patient did not exhibit any cutaneous manifestations, which are commonly observed in BPDCN, highlighting the atypical presentation in this case. Cytogenetic analysis remained normal. NGS revealed mutations in JAK2 (1.6%), DNMT3A (13.8%), TET2 (13%), IKZF1 (9.2%), and MPL (3.9%). Comparative genetic analysis revealed that the founding ​DNMT3A​ and ​TET2​ mutations persisted, while the clone had newly acquired mutations in ​JAK2, ​MPL, and ​IKZF1, illustrating a dynamic clonal evolution (Figure 1E). Although lumbar puncture was considered to evaluate potential CNS involvement, the procedure was declined by the patient’s family. Ultimately, the patient and family decided to forego further treatment, leading to the patient’s discharge home.

This case provides compelling evidence of a shared clonal origin between AML and secondary BPDCN, as demonstrated by the persistence of ancestral TET2 and DNMT3A mutations. Our literature review, in addition to the present case, identified 9 reported cases of BPDCN with preceding or concomitant hematologic malignancies and clearly documented clonal data, with involved sites at diagnosis encompassing the cutis, BM, and CNS (Table 1). Among previously reported cases of PCHM, 3 were simultaneous occurrences (2 MDS, 1 CMML), and 6 were previous events (4 CMML, 2 MDS). Of these previous events, 3 cases progressed to AML, with 1 case occurring after BPDCN. Among the 9 patients, except for one patient with MDS who did not show any detectable mutational clones, the remaining 8 patients all shared the TET2 mutation between BPDCN and PCHM. Notably, 3 of these patients had early clonal hematopoiesis data available, and one patient had previously been diagnosed with Anaplastic Lymphoma Kinase 1-negative Anaplastic Large Cell Lymphoma (ALK1-ALCL), with the shared TET2 clone present between all these conditions. As far as we know, this report presents the first case of de novo AML progressing to BPDCN, confirmed through shared mutations.

TET2, which encodes a DNA dioxygenase essential for epigenetic regulation, is frequently mutated in clonal hematopoiesis (CH) (6). Griffin et al. (7) demonstrated that plasmacytoid dendritic cells harboring TET2 mutations exhibit resistance to UV-induced apoptosis, implicating this mutation in the pathogenesis of BPDCN. These findings suggest that TET2-driven epigenetic dysregulation may serve as a shared molecular foundation underlying divergent hematologic malignancies, warranting further mechanistic investigation. Although JAK2 and MPL mutations have not been previously linked to BPDCN, IKZF1 alterations have been reported in rare cases, suggesting a potential role in its pathogenesis that warrants further investigation (8).

Clonal hematopoiesis of indeterminate potential (CHIP) serves as a critical precursor to leukemogenesis, characterized by recurrent mutations—most frequently in DNMT3A, TET2, and ASXL1—that emerge in otherwise healthy individuals and significantly elevate the risk of AML and other myeloid neoplasms (9, 10). Foundational research confirms that preleukemic hematopoietic stem cells (HSCs) carrying DNMT3A and TET2 mutations can survive therapy and give rise to diverse malignant lineages (9, 10). In AML, large-scale genomic studies reveal substantial biological heterogeneity, with multiple subclones often coexisting at diagnosis and undergoing branched evolutionary trajectories at relapse (11, 12), frequently influenced by chemotherapy-induced selective pressures that reshape clonal architecture. In BPDCN, the high prevalence of marrow CH and the frequent sharing of TET2 mutations with associated myeloid malignancies support a model in which CH-derived ancestral clones undergo lineage-specific transformation driven by microenvironmental cues and secondary mutational events (5), consistent with the mechanism of UV-induced pDC leukemic transformation observed in the skin (7). In the present case, the disappearance of detectable TP53 at relapse despite persistent TET2/DNMT3A mutations suggests that relapse originated from a preleukemic ancestral clone rather than representing continued evolution of the initial TP53-mutant subclone, aligning with established models of AML clonal heterogeneity and evolution (11). Although TP53 mutations are generally associated with poor prognosis and often persist, their subclonal distribution and potential for eradication under therapeutic pressure have been documented (13).

Despite the common basis of TET2 mutations, the hierarchical relationships of clonal evolution in malignant clones during disease progression remain unknown. The mechanisms underlying the shared clonal origin of BPDCN and PCHM may involve two distinct pathways: divergent evolution from a common CH progenitor or transformation of BPDCN from an underlying myeloid neoplasm. Mature plasmacytoid dendritic cell proliferation (MPDCP) was officially mentioned only in WHO-5 and is associated with a myeloid neoplasm, frequently CMML, but also MDS or AML, particularly with monocytic differentiation (14–19). Another interesting phenomenon is that CMML cases have also been found to harbor plasmacytoid dendritic cell nodules (20). Therefore, BPDCN may potentially arise from these pDCs, which have already undergone early overproliferation or even clonal expansion, and further deteriorate due to a secondary hit. However, most institutions gave insufficient attention to flow cytometry analysis of pDCs in myeloid malignancies, resulting in limited exploration of the association between myeloid neoplasms and pDCs, and leaving the nature of this correlation still unclear. Therefore, when CMML patients or those with TET2 mutations present with clinical findings inconsistent with disease relapse, performing pDC analysis may help further elucidate the mechanisms underlying the development of secondary BPDCN. Future studies with larger cohorts are warranted to further elucidate the roles of TET2 and DNMT3A in the pathogenesis and clonal evolution of BPDCN.

We report a case of BPDCN secondary to de novo AML, characterized by shared TET2 and DNMT3A mutations. Our literature analysis underscores the pivotal role of TET2 mutations as a common genetic driver in the pathogenesis of BPDCN arising from myeloid neoplasms. Routine examination of pDCs in myeloid malignancies, particularly those harboring TET2 mutations, may be essential for elucidating the mechanisms underlying BPDCN development and its clonal evolution.