Version July 2025
INTRODUCTION —
Acute promyelocytic leukemia (APL) is a clinically and biologically distinct subtype of acute myeloid leukemia (AML) that is characterized by the infiltration of bone marrow by malignant promyelocytes. APL frequently presents with a coagulopathy that requires urgent management. APL was called AML-M3 in the now obsolete French-American-British (FAB) classification system for AML.
APL is characterized by genetic rearrangement between the gene that encodes retinoic acid receptor alpha (RARA) on chromosome 17 and a translocation partner on another chromosome. Nearly all cases of genetically confirmed APL involve rearrangement of RARA with the PML gene on chromosome 15. The typical syndrome, APL with t(15;17)(q24.1;q21.2)/PML::RARA, is sensitive to treatment with all-trans retinoic acid (ATRA) and arsenic trioxide (ATO). Responsiveness to ATRA and ATO by APL with t(15;17) is unique among all subtypes of AML.
Rare variants of APL involve the rearrangement of RARA with partner genes on other chromosomes (eg, chromosomes 11 or 5). Many of these APL variants are resistant to ATRA.
This topic discusses the molecular biology of APL.
Related topics include:
●(See "Initial treatment of acute promyelocytic leukemia in adults".)
●(See "Acute myeloid leukemia: Classification".)
RETINOIC ACID RECEPTORS —
Retinoic acid receptor (RAR) refers to a group of nuclear hormone receptors that mediate growth, differentiation, and survival in various tissues through binding with its ligand, retinoic acid (RA).
RAR alpha (RARA) is one of three types of RAR; the others are RAR beta (RARB) and RAR gamma (RARG) [1]. RARA is the primary form of RAR expressed in hematopoietic cells [2]. RAR transcriptional regulation is important for hematopoietic stem cell (HSC) development, maintenance, and expansion, as well as for their maturation and differentiation into distinct hematopoietic cell lineages.
RARA contains discrete functional domains, including an amino-terminal transcriptional activation domain, along with deoxyribonucleic acid (DNA)-binding, dimerization, and retinoid-binding domains. RARA heterodimerizes with retinoid X receptor (RXR), and the heterodimer binds to RA-responsive elements (RAREs) on DNA to regulate transcription of target genes [2]. Retinoids were known to have important roles for differentiation of hematopoietic cells and other tissues long before their interactions with RARs were understood.
Ligand-free RARA binds RAREs and represses transcription of RA-responsive genes through its interactions with corepressor molecules, including NCOR and SMRT. These interactions trigger recruitment of a high molecular weight complex containing histone deacetylase (HDAC) to maintain chromatin in a densely packed, inactive state [2]. The binding of RA to RARA dissociates NCOR from the RARA/RXR heterodimer, relieving transcriptional repression and thereby activating genes that lead to terminal differentiation of myeloid cells (picture 1A-B) [3,4].
t(15;17), THE USUAL TRANSLOCATION IN APL —
The leukemic cells of >90 percent of patients with APL have a balanced translocation t(15;17)(q24.1;q21.1), which involves retinoic acid receptor alpha (RARA) on chromosome 17 and promyelocytic leukemia (PML) on chromosome 15 [5,6].
An additional 5 percent of patients express a PML::RARA fusion gene, but they do not have the classic t(15;17). These cases are caused by insertions or other complex chromosomal rearrangements [7].
PML — The PML protein participates in various cellular processes involving homeostasis and tumor suppression.
PML is involved with cellular responses to viral infections, stress, senescence, angiogenesis, differentiation, and maintenance of genome stability. PML appears to act by organizing cellular organelles called nuclear bodies (PML-NBs) [8]. PML-NBs have roles in DNA damage response (DDR) and self-renewal of normal and cancer stem cells [9], but the precise mechanisms by which PML and NBs exert cellular responses are not well-defined.
PML, the gene that encodes PML, was first identified through its involvement with RARA in t(15;17) [10,11]. Six nuclear isoforms and one cytoplasmic isoform of PML are generated by alternative splicing of C-terminal exons [2].
PML::RARA — The t(15;17) chromosomal rearrangement creates a PML::RARA fusion gene on the der(15) chromosome that deregulates transcriptional control and disrupts PML homeostatic function.
PML::RARA encodes a protein in which the DNA-binding and dimerization domains of PML are fused to C-terminal portions of RARA, including the RA binding site. PML::RARA acts as a transcriptional repressor of RARA target genes that leads to the proliferation of myeloid progenitors and maturation arrest at the promyelocytic stage, and disrupts PML-NBs [12,13]. The PML::RARA phenotype is also characterized by the downregulation of autophagy-related genes, and altered senescence, apoptosis, and DNA repair [14-19].
Retinoid binding induces conformational changes that cause the dissociation of corepressors and the induction of transcriptional activation of responsive genes [1]. PML::RARA binds RA-responsive elements (RAREs) and exerts a dominant negative action on transcription, inhibiting activation by physiologic ligands and causing the maturation arrest at the promyelocyte stage (figure 1).
Clinical and pathologic features of APL are presented separately. (See "Clinical manifestations, pathologic features, and diagnosis of acute promyelocytic leukemia in adults".)
RARA::PML — The role of the reciprocal RARA::PML fusion on the der(17) chromosome is uncertain.
RARA::PML is detected in approximately 80 percent of cases of t(15;17) [20], but its contribution, if any, to the development of APL is unclear.
Additional genetic events — Although PML::RARA is the primary molecular driver, other molecular events may be involved in the development of APL.
Whole-exome sequencing of 12 cases of de novo APL reported recurrent alterations of FLT3 (43 percent), WT1 (14 percent), NRAS (10 percent), and KRAS (4 percent), whereas mutations in other genes that commonly mutated in various subtypes of acute myeloid leukemia (AML) and myeloid malignancies (eg, DNMT3A, NPM1, TET2, ASXL1, IDH1/2) were notably absent [21]. Frequent FLT3 alterations have also been described in other studies of APL [2].
In one study, FLT3-internal tandem duplication (ITD) was found in 20 percent of 171 patients with APL who were treated with all-trans retinoic acid (ATRA) plus anthracycline-based chemotherapy [22]. Compared with wild-type FLT3, FLT3-ITD was associated with inferior three-year overall survival (OS; 62 versus 82 percent; hazard ratio [HR] 2.39 [95% CI 1.17-4.89]) in multivariable analysis. In another study, FLT3-ITD was reported in 31 percent of 245 patients with APL [23]. FLT3-ITD did not appear to impact remission rate, induction death rate, disease-free survival, or OS in patients treated with ATRA plus anthracycline-based chemotherapy, with or without arsenic trioxide (ATO) consolidation.
Other genetic mutations may also contribute to disease pathogenesis. As an example, whole-exome sequencing of 30 patients with newly diagnosed APL demonstrated FLT3 mutations and alterations in STAG2, U2AF1, SMC1A, USP9X, IKZF1, LYN, MYCBP2, and PTPN11, which have been implicated in other myeloid malignancies [24].
Animal models also support the concept that additional mutations contribute to the development of APL. Transgenic expression of PML::RARA causes a myeloproliferative condition that resembles chronic myeloid leukemia [25,26]. However, the emergence of full-blown APL requires >12 months, which suggests that additional genetic/epigenetic changes may be necessary to induce APL in mice. JAK signaling and/or epigenetic changes have been implicated in this process [27].
VARIANT TRANSLOCATIONS —
Rare patients with APL present with variant translocations, in which the retinoic acid receptor alpha (RARA) is rearranged with a partner gene other than PML.
An international registry reported that 0.4 percent of 2895 patients with genetically confirmed APL had a variant translocation of RARA [28]. A review of the literature reported 14 different variant translocations. The incidence of variants in this study is lower than previous estimates of 1 to 2 percent. The small number of cases limited the interpretation of clinical differences in variants compared with conventional APL with t(15;17)/PML::RARA. The most common variants included partner genes on chromosomes 11 or 5.
PLZF::RARA and t(11;17) — The translocation t(11;17)(q23;q21.1), which is associated with the rearrangement of RARA with promyelocytic leukemia zinc finger (PLZF), is the most common variant rearrangement in APL.
In one study, t(11;17) was reported in 5 patients among 611 cases of APL [7], but a larger registry study reported it in only 9 of 2895 cases [28].
Most of the 35 reported cases of APL with PLZF::RARA were in males, with a median age of 48 years at presentation [28]. APL associated with PLZF::RARA may have distinct morphologic features. By contrast with APL with t(15;17), the t(11;17) variants are more likely to have a conventional (rather than bilobed) nucleus, cytoplasmic granules that are fine or absent, and increased expression of CD56 by flow cytometry [29-31].
Most cases of APL with PLZF::RARA appeared to be resistant to all-trans retinoic acid (ATRA) [28]. However, two-thirds of patients achieved complete remission (CR) with either first-line or salvage induction therapy, using various combinations of ATRA, acute myeloid leukemia (AML)-type chemotherapy, and/or arsenic trioxide (ATO).
In these tumors, the 3' end of RARA is fused to the 5' end of PLZF, which encodes a protein containing nine zinc fingers, a motif frequently found in transcription factors [29-31]. PLZF is expressed in myeloid, but not lymphoid, lineages; its expression is downregulated during differentiation. PLZF::RARA antagonizes the normal function of RARA::RXRA (retinoid X receptor alpha) heterodimers, suggesting that it behaves in a dominant negative manner. Unlike PML, PLZF is not a component of nuclear bodies (NBs); instead, it is localized in smaller, more numerous nuclear subdomains.
NPM1::RARA and t(5;17) — A rare variant translocation in APL with t(5;17) has been described.
APL with t(5;17)(q35;q21.1) is associated with fusion of NPM1, which encodes a nucleolar phosphoprotein, with RARA [7,28,32].
Among 9 cases of APL with NPM1::RARA, 56 percent were in children, and the median age at diagnosis was 12 years; 6 cases were in males [28]. Treatment with ATRA, with or without chemotherapy-based induction, was associated with 78 percent CR; responsiveness to ATO is not described.
NPM1 is also involved in t(2;5)(p23;q35) in anaplastic large cell lymphoma, where it fuses to ALK (anaplastic lymphoma kinase). NPM1 has been found fused to MLF1 in t(3;5)(q25.1;q35) in AML.
Other rare variants — Other rare APL variants [28] include:
●STAT5B::RARA/interstitial chromosome 17 deletion [33-35]
●TBLR1::RARA/t(3;17)(q26;q12-21) [36,37]
●NuMA::RARA/t(11;17)(q13;q21.1) [6]
●BCOR::RARA/t(X;17)(q11;q21.1) [38]
SUMMARY
●Description – Acute promyelocytic leukemia (APL) is a biologically and clinically distinct subtype of acute myeloid leukemia (AML) characterized by a translocation involving the RARA (retinoic acid receptor alpha) gene on chromosome 17. APL was formerly called AML-M3. Most cases of APL are sensitive to treatment with all-trans retinoic acid (ATRA) and arsenic trioxide (ATO).
●Retinoic acid receptors – Retinoic acid receptors (RARs) are nuclear hormone receptors that mediate the differentiation and function of various tissues through binding with retinoic acid (RA). RARA is the primary RAR expressed in hematopoietic cells. (See 'Retinoic acid receptors' above.)
RARs heterodimerize with retinoid X receptor (RXR), and the heterodimer binds RA response elements (RAREs) in target genes, thereby regulating myeloid gene expression. The RARA/RXR heterodimer interacts with the nuclear corepressor, NCOR, a ubiquitous nuclear protein that mediates transcriptional repression, but the binding of RA to RARA releases this repression.
●Usual APL translocation – More than 90 percent of APL cases have the balanced translocation t(15;17)(q24.1;q21.1) that involves RARA on chromosome 17 and promyelocytic leukemia (PML) on chromosome 15. (See 't(15;17), the usual translocation in APL' above.)
Approximately 5 percent of APL cases express the PML::RARA fusion gene, but they do not have the classic t(15;17) chromosomal translocation; these cases are caused by insertions or other complex chromosomal rearrangements.
●PML – PML is a ubiquitously expressed gene found in organelles called nuclear bodies (PML-NBs). PML is involved with homeostasis and tumor suppression, but its mechanisms of action are not well-defined. (See 'PML' above.)
●PML::RARA – t(15;17) can create two fusion genes (see 'PML::RARA' above):
•PML::RARA on the der(15) chromosome is found in all cases of t(15;17). This rearrangement encodes a fusion protein in which the DNA-binding and dimerization domains of PML are fused to C-terminal portions of RARA, including the retinoid binding site. PML::RARA functions as an aberrant retinoid receptor that blocks RA-induced myeloid differentiation.
PML::RARA may act, in part, through epigenetic changes in target genes. However, the precise mechanisms by which it causes APL and the essential target genes are not well-defined.
•The reciprocal translocation generates RARA::PML on the der(17) that can be detected in 80 percent of cases.
●APL variant translocations – Rare variant translocations that cause APL have been described. Detection of these rare variants may have therapeutic implications for the treatment of APL, as many have little or no responsiveness to ATRA (see 'Variant translocations' above):
•PLZF::RARA – APL with t(11;17)(q23;q21.1) is associated with the fusion of RARA with promyelocytic leukemia zinc finger (PLZF). (See 'PLZF::RARA and t(11;17)' above.)
•NPM1::RARA – APL with t(5;17)(q35;q21.1) is associated with the fusion of RARA with NPM1 (nucleophosmin). (See 'NPM1::RARA and t(5;17)' above.)
•Other rare variants – Rare cases of APL are associated with the rearrangement of RARA with STAT5B, TBLR1, NuMA, BCOR, and other genes. (See 'Other rare variants' above.)
