Acute promyelocytic leukemia (APL) accounts for approximately 10-15% of newly diagnosed acute myeloid leukemia (AML) cases, associated with life-threatening early bleeding and death [1,2,3,4]. The introduction of all-trans retinoic acid (ATRA) and arsenic trioxide (ATO) has transformed APL from a highly fatal disease to a highly curable one. However, early death rates ranging from 8.2 to 32.6% remain a significant concern [3,4,5,6]. Therefore, prompt diagnosis and immediate initiation of appropriate therapy are critical.

The diagnostic hallmark of APL is the balanced reciprocal translocation t(15;17)(q24.1;q21.2) involving the promyelocytic leukemia (PML) gene located on chromosome 15q24.1 and the retinoic acid receptor alpha (RARα) gene located on chromosome 17q21.2, leading to the formation of the PML::RARα fusion gene [7,8,9]. The PML-RARA oncoprotein impairs normal myeloid differentiation and maturation, alters gene expression, deregulates transcriptional control, and inhibits apoptosis, playing a central role in the pathogenesis of APL and the differentiation response to ATRA [7,8,9].

Less than 10% of APL patients lack the classical t(15;17), but still harbor the PML::RARα fusion gene, as a result of insertional (cryptic or masked) events, or simple or complex variant translocations, involving chromosomes 15, 17 together with one or more additional chromosome(s) [10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. These rare PML::RARα rearrangements are difficult to diagnose, and their rapid identification is crucial for selecting the most effective treatment.

We report a previously undescribed complex variant translocation involving chromosomes 8, 15 and 17 in a patient with APL, who exhibited an unexpected typical fluorescence in situ hybridization (FISH) pattern and was negative by reverse transcription-polymerase chain reaction (RT-PCR).

A 59-year-old man with no significant past medical history was referred to our institution in May 2025 for the evaluation and treatment of pancytopenia, detected on a routine blood test. Complete blood count revealed Leukopenia (White Blood Cells 1.4 × 10⁹/L, Neutrophils 15%, Monocytes 14%, Lymphocytes 71%), Anaemia (Hb 9.3 g/dL, MCV 97.6 fL, MCH 36.6 pg), and Thrombocytopenia (PLT 20 × 10⁹/L). The patient was asymptomatic and clinical examination was negative for lymphadenopathy or organomegaly. On admission, he underwent complete laboratory testing, peripheral blood smear examination, bone marrow aspirate, bone marrow immunophenotyping by Flow Cytometry, and cytogenetic analysis. In addition, DNA and RNA were extracted from the bone marrow sample for prospective molecular testing. Coagulation parameters were as follows: Prothrombin Time (PT) 14 sec (9.0-13.0), Activated Partial Thromboplastin Time (aPTT) 30.02 sec (26.00-38.00), Fibrinogen 2.33 g/L (2.00-4.50). Renal function, liver tests, and lactate dehydrogenase (LDH) were within normal limits. Microscopic examination of the peripheral blood smear revealed the presence of rare atypical blastoid cells. Bone marrow aspirate was positive for the presence of a uniform blast cell population characterized by large cell size, intermediate nucleus-to-cytoplasm ratio, kidney-shaped nuclei, and coarse granules and occasionally Auer rods (Figure 1A). These morphological findings were indicative of APL. Consequently, molecular testing was requested specifically targeting the PML::RARα fusion transcript.

Bone marrow immunophenotyping by flow cytometry revealed a homogeneous population of intermediate-sized cells with increased granularity/intracellular complexity and immunophenotypic characteristics CD45⁺CD34^-CD117⁺CD13^hetCD33⁺⁺HLA-DR^-MPO⁺⁺CD11b^-CD9⁺. The combination CD9⁺CD11b^-HLA-DR^- is considered to have 85% sensitivity and 95% specificity for classic APL, further confirming the suspicion of APL (Figure 1B). Per clinical guidelines, ATRA treatment was initiated immediately. Despite the negative molecular result for all three isoforms of the PML::RARα fusion transcript (L-bcr1, V-bcr2, S-bcr3), ATRA therapy was continued pending cytogenetic results because of the strong morphological and immunophenotypic evidence supporting the diagnosis of APL.

Conventional cytogenetic analysis was performed on 24 h and 48 h unstimulated bone marrow cultures using GTG (trypsin-Giemsa) banding at the time of his admission. Κaryotypes were described according to the 2024 International System for Human Cytogenomic Nomenclature [37]. Among 24 metaphases analyzed, eight had a normal karyotype (46,XY), whereas 16 had a complex karyotype involving the long arms (q) of chromosomes 15 and 17 and the short arm (p) of chromosome 8 (Figure 2).

Dual-color, dual-fusion (D-FISH) analysis was performed on 24 h unstimulated cells using the t(15;17) PML/RARA probe (Μetasystems). Fluorescent signals were evaluated on metaphases and in at least 200 nuclei. Two PML/RARA fusion signals were detected in 166/200 (83%) interphase nuclei and in 5 metaphases analyzed, confirming the presence of the t(15;17)(q24;q21.2) translocation (Figure 3A, 3B). One PML::RARα fusion signal was located on the derivative chromosome der(17), whereas the second signal was detected not on the der(15), but on a larger chromosome (Figure 3A, 3B). Using XCE 8 enumeration and t(8;9)(p22;p24) PCM1/JAK2 dual-color, dual-fusion probes (Metasystems), two 8 chromosome centromeres, two 8p22 signals, and extra chromosomal material on one chromosome 8p were observed (Figure 3C, 3D).

Finally, the diagnosis of “Acute Promyelocytic Leukemia with PML::RARα fusion” was established. The patient was treated according to the PETHEMA LPA 2017 protocol (PETHEMA/PALG, Treatment of Acute Promyelocytic Leukemia, PETHEMA LPA 2017) and was classified as a low/intermediate-risk patient (WBC ＜10 × 10⁹/L, age ＜70 years). However, due to a lack of ATO (Arsenic Trioxide), the patient followed the schedule of high-risk (≥60 years) patients. He started induction therapy with ATRA (45 mg/m²/d) and Idarubicin (12 mg/m²) on days 2, 4, 6, and 8. Triple consolidation therapy followed: course 1 with ATRA (45 mg/m²/d × 15 days) and Idarubicin (5 mg/m²/d × 4 days); course 2 with ATRA (45 mg/m²/d × 15 days) and Idarubicin (8 mg/m² × 3 days); course 3 with ATRA (45 mg/m²/d × 15 days) and Idarubicin (12 mg/m²/d × 1 day), after the completion of which, monitoring of the disease was mandated by the protocol. Figure 4 provides a timeline of the treatments.

During induction, the patient had mild differentiation syndrome treated with corticosteroids. No cardiotoxicity was observed during the treatment period. Febrile neutropenia episodes occurred during the hematological toxicity period of each scheduled treatment scheme, but no evidence of bacteremia was detected.

In October 2025, the patient was in complete hematological, morphological, immunophenotypic and cytogenetic response. A repeat FISH analysis on bone marrow biopsy was negative for APL. In January 2026, the patient remained in complete hematological, morphological, immunophenotypic and cytogenetic response, and repeat FISH analysis was negative. The next remission assessment is pending.

The formation of the PML::RARα fusion gene is the defining molecular event in APL, irrespective of whether it arises from the classic t(15;17) translocation or from atypical/variant rearrangements, or of its genomic location. Consequently, accurate identification of the fusion is of great importance for selecting the most effective and potentially life-saving treatment strategy [15,16]. Currently, APL diagnosis is based on PML::RARα detection through immunophenotyping, conventional karyotyping, fluorescent in situ hybridization (FISH), real-time quantitative reverse transcription polymerase chain reaction (RT-PCR), and next-generation sequencing (NGS). However, atypical/variant cases resulting from occult t(15;17) rearrangements remain diagnostically challenging and may remain undetectable.

Several atypical cases lacking the classic translocation t(15;17) have been described in APL, involving complex variant translocations that include chromosomes 15 and 17 together with one additional chromosome (three-way translocations) or more chromosomes (four-way translocations). Most autosomes, except chromosomes 14 and 21, and the X chromosome, have been reported to participate in these rearrangements [10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. Although there is no obvious clustering to particular chromosomal bands, the implication of putative oncogenes located in the breakpoint regions in the pathogenesis of APL is suggested [10,11,17,18,19,20,21,22,23,24,25,34,35]. According to the literature, patients with atypical APL exhibit comparable clinical features to those of typical APL, with an approximately equal male-to-female ratio and a median age at diagnosis of 45 years [38].

We present a patient with APL harboring a complex three-way translocation between chromosomes 8p23, 15q24 and 17q21. The involvement of chromosome 8 in three-way translocations associated with APL has been previously described, with several breakpoints implicated (p23, q11.2, q21.2, q22 and q24) [15,22,29]. The interesting thing in our case is the formation of a complex three-way translocation involving 4 breaks, two very closely spaced breakpoints on chromosome 15, one on chromosome 17 and one on chromosome 8. It is assumed that first, the typical, critical for APL, reciprocal translocation between 15q24 and 17q21 occurred, leading to the formation of two PML::RARα fusion signals on der(15) and der(17), followed by a second reciprocal translocation between der(15) with an additional breakpoint at 15q22, and chromosome 8p23, resulting in the formation of der(8) (Figure 5). Although the actual order of events cannot be confirmed, the observation of a single neoplastic clone carrying only the three-way translocation, without any detectable cells harboring the classical t(15;17), raises the possibility that the rearrangements occurred simultaneously rather than sequentially [16,24].

The above findings were confirmed by D-FISH analysis, which showed the typical signal pattern of one red, one green, and two fusion signals. One fusion signal was located on the der(17), whereas the second was detected on chromosome 8 rather than on der(15), due to the translocation between these two chromosomes. In most cases with complex variant translocations t(15;17), an atypical D-FISH pattern consisting of one fusion, two red and two green signals, usually results from a 3-point three-way translocation involving der(17) and a third chromosome [18]. One red and one green signal remain on the normal chromosomes 15 and 17, respectively; the single PML::RARα fusion signal is detected on der(15), while the reciprocal fusion signal is absent, because the corresponding red and green signals are separated and located on der(17) and the third chromosome (Table 1). Kim et al. (2023) reported a case in which one fusion signal was identified on der(17) and not on der(15), due to a complex three-way translocation involving 15q14, 15q24, and 17q21 [30]. Regarding the cases involving chromosome 8, D-FISH analysis was performed in four out of five patients, showing PML::RARα fusion signal on der(15), red signals on der(8) and the normal chromosome 15, and green signals on der(17) and the normal chromosome 17 [15,22].

To our knowledge, only a few cases exhibiting a typical D-FISH pattern have been reported to date (Table 1). Fujishima et al. (2000) and Bennour et al. (2013) described cases in which one PML::RARα fusion signal was abnormally located on der(2) and der(12), respectively, rather than on der(15) [16,24]. Wan et al. (2007) showed a four-break, three-way translocation in which two PML::RARα fusion signals were detected on der(15) and der(7) [18]. Gagnon et al. (2022) reported two additional cases with a typical D-FISH pattern; however, no further information was provided [29].

Although both classical (karyotyping) and molecular (FISH) cytogenetics confirmed the presence of the PML::RARα fusion, RT-PCR was negative. In only two of five previously reported cases of three-way translocations involving chromosome 8 was RT-PCR performed, both were positive for PML::RARα transcripts [15,22]. RT-PCR is a highly sensitive technique for detecting the three typical PML::RARα transcripts, due to different breakpoints within the PML gene: the long (L, or bcr1) transcript resulting from a breakpoint in intron 6, the variant (V, or bcr2) transcript involving exon 6, and the short (S, or bcr3) transcript resulting from a breakpoint in intron 3, involving 55%, 5%, and 40% of all t(15;17)-positive APL cases, respectively [38,39,40,41]. Rarely, variant PML::RARα translocations, insertions of genomic DNA sequences at the junction of the PML and RARα genes, or partial deletions of the RARα or PML genes, lead to atypical PML::RARα fusions, not readily identifiable using conventional primers [38,39,40,41]. These unknown breakpoints result in different PML::RARα transcripts, and their characterization and sequencing are important for understanding pathogenic molecular mechanisms and monitoring minimal residual disease (MRD). In our patient, a positive FISH result indicated the presence of the PML::RARα fusion gene, but RT-PCR was false-negative due to an atypical, novel breakpoint in the PML gene.

Our patient achieved complete hematological, morphological, immunophenotypic, and cytogenetic responses after treatment, suggesting that the abnormal localization of the PML::RARα fusion gene on chromosome 8 did not interfere with the clinical response to therapy. Cases with both typical and atypical PML-RARA isoforms have been correlated with diverse prognosis and responsiveness to treatments (e.g., ATRA), likely due to the different PML domains retained in the fusion protein [42]. However, studies have shown that the presence of the PML::RARα gene is crucial for achieving the best treatment response, regardless of its genomic location. No difference has been reported between variant and typical APL cases and according to the most recent recommendations, standard therapy should not be changed based on the PML-RARA isoforms [16,17,18,19,23,24,35,42].

Although traditional diagnostic techniques, such as karyotyping, FISH and RT-PCR, provide valuable information, they all have the limitation of incomplete genomic coverage [43]. NGS technology has fundamentally transformed the fields of genomics and molecular diagnostics by enabling rapid, cost-effective genetic analysis through methodologies, such as targeted gene panels, Whole Exome Sequencing (WES), and Whole Genome Sequencing (WGS) [43]. Unlike traditional sequencing methods, which are labor-intensive and limited in scope, NGS-based approaches enable the simultaneous sequencing of millions of DNA fragments and the comprehensive examination of the genetic landscape of leukemia [43]. NGS can detect a wide range of genetic alterations and has revolutionized the diagnosis, treatment, and minimal residual disease monitoring in leukemia [43]. However, high cost, complex data interpretation, and the lack of global standardization are limitations that currently prevent its universal adoption in clinical practice [43]. The final decision on conducting NGS in our case was driven primarily by financial constraints, as insurance providers do not cover this test.

Our findings demonstrate the first case of APL harboring a complex variant three-way translocation involving chromosome 8p23. The difficulty of identifying the PML::RARα fusion gene in APL patients with variant t(15;17) rearrangements involving uncommon breakpoints highlights the importance of an integrated diagnostic approach combining morphology, immunophenotyping, classical (karyotyping) and molecular (FISH) cytogenetics, for optimal management of APL patients and better understanding of the pathogenesis of the disease. Further sequencing studies of both typical and atypical APL patients might shed light on the pathophysiology of the disease, the responsiveness to treatment and may aid in the development of novel targeted therapies.