The PRKAR1A gene is fused to RARA in a new variant acute promyelocytic leukemia

The majority of acute promyelocytic leukemia (APL) cases are characterized by the PML-RARA fusion gene, usually as a consequence of the t(15;17)(q22;q21) translocation.¹  As a result, retinoid sensitivity of the retinoic acid receptor α (RARα), which normally functions as a retinoid-inducible transcription factor, is reduced, causing a block in myeloid differentiation.²  Prolonged disease-free survival can be achieved by combining all-trans retinoic acid (ATRA) and chemotherapy.³  Arsenic trioxide (As₂O₃) is of proven value in relapsed disease⁴  and is also effective during initial induction or consolidation or both.⁵  In a small number of APL variants, RARA (at chromosome 17q21) is fused with an alternative partner gene; promyelocytic leukemia zinc finger (PLZF, 11q13),⁶  nucleophosmin (NPM, 5q35),⁷  nuclear mitotic apparatus (NUMA, 11q23),⁸  or signal transducer and activator of transcription 5b (STAT5B, 17q21).⁹  RARα fusions to PML, NPM, and NUMA are ATRA responsive, whereas PLZF-RARα is ATRA resistant.^10,11  This report describes a new variant APL and identifies the RARA partner gene.

The patient concerned in this report was originally registered in a multicenter national trial for acute promyelocytic leukemia that had institutional ethics approval at the hospital concerned. Informed consent was obtained in accordance with the Declaration of Helsinki. The patient was subsequently deemed ineligible for that trial because he failed to meet the required genetic inclusion criterion. This report is concerned with an in vitro characterization of the patient's leukemic cells and did not involve any direct patient experimentation.

A 66-year-old man with a history of excessive alcohol consumption, clinical evidence of chronic liver disease, and a 14-month history of mild thrombocytopenia was investigated for lethargy, anorexia, and weight loss. A full blood count showed a hemoglobin level of 104 g/L, white cell count of 5.3 × 10⁹/L, neutrophil count of 2.1 × 10⁹/L, and platelet count of 93 × 10⁹/L. The blood film was leukoerythroblastic with occasional hypergranular promyelocytes. Both creatinine (0.22 mmol/L) and γ-glutamyltransferase (5.17 μkat/L [310 U/L]) were mildly elevated. Coagulopathy was absent apart from mildly increased D-dimers (0.4 mg/L; normal, < 0.2 mg/L). A markedly hypercellular marrow contained 88% hypergranular promyelocytes, and most exhibited regular nuclear outlines (Figure 1A). Auer rods and faggot cells were absent. Flow cytometric analysis revealed strong intracellular myeloperoxidase expression; however, expression of CD13, CD33, and CD11b was weak. The cells were negative for CD2, CD19, CD34, CD56, CD117, and HLA-DR. The karyotype was 47,XY,+22[5]/46,XY[30], without the classic t(15;17) translocation. Fluorescence in situ hybridization (FISH) studies showed del(17)(q21)(5′RARA-,3′RARA+) (Figure 1B) plus a split RARA signal (Figure 1C) with RARA break-apart and PML-RARA dual fusion probes, respectively.

Treatment was commenced with ATRA (45 mg/m²/d), idarubicin (9 mg/m²/d on days 2, 4, 6, and 8), and As₂O₃ (0.15 mg/kg/d from day 9); however, arsenic was suspended on day 22 because of sepsis and cardiac toxicity. Complete remission was documented on day 35 by morphology, cytogenetics, and interphase FISH. Two cycles of consolidation (infusional cytarabine 100 mg/m²/d × 7 and amsacrine 100 mg/m²/d × 3) were administered, superimposed on continuous ATRA, and were followed by maintenance therapy (ATRA for 2 weeks every 3 months). Eleven months after diagnosis, a full blood count showed mild anemia (99 g/L), normal neutrophils, and marked thrombocytopenia (20 × 10⁹/L), and the marrow showed essentially normal trilineage hemopoiesis. Cytogenetic analysis and interphase FISH were both normal (46,XY[40] and nuc ish(RARAx2)[72], respectively). Two years from diagnosis, the patient remains clinically free of leukemia with further hematologic improvement: normal hemoglobin level, normal neutrophil count, and moderate thrombocytopenia (platelets, 61 × 10⁹/L).

The combination of APL morphology and RARA disruption in the absence of PML involvement by FISH suggested the possibility of a novel translocation in the pathogenesis of this leukemia. A classic PML-RARA fusion was absent when PML- and RARA-specific primers were used (data not shown). However, an assay for bcr3 transcripts generated 2 products with high-quality sequence identity to exons 2 and 3 of the PRKAR1A gene, followed by the 5′ end of RARA exon 3 (Figure S1, available on the Blood website; see the Supplemental Materials link at the top of the online article). PRKAR1A encodes the regulatory subunit type I-α (RIα) of cyclic adenosine monophosphate (cAMP)–dependent protein kinase (PKA).¹² 

To confirm the presence of the suggested PRKAR1A-RARA fusion (Figure S2), nested reverse transcription–polymerase chain reaction (RT-PCR) was performed with PRKAR1A-specific forward primers and reverse primers that would amplify substantial amounts of RARA (Figure 2A). Both first- and second-round PCR products of expected sizes were obtained (Figure 2B,C) and confirmed to be PRKAR1A-RARA fusion transcripts by DNA sequencing (Figure 2D). In contrast, RT-PCR for PRKAR1A-RARA was negative in the 11-month remission sample (Figure 2C).

FISH, using a RARA probe and a BAC probe (RP11–120M18) encompassing the PRKAR1A gene, confirmed the presence of a PRKAR1A-RARA fusion on der(17) (Figure 1D). The simplest explanation would be an insertion of RARA distal to PRKAR1A followed by a deletion removing 3′ PRKAR1A, 5′ RARA, and any intervening sequences.

Alternative splicing of PRKAR1A explains the presence of 2 PRKAR1A-RARA PCR products observed in the diagnostic sample. The longer transcript results from cryptic splicing of the first 100 bases of PRKAR1A exon 3 to RARA exon 3, producing an in-frame fusion transcript capable of encoding a 495–amino acid RIα-RARα fusion protein. Translation of the putative open reading frame would generate a protein containing the RIα protein dimerization domain fused to the same carboxy terminal end of the RARα protein shared by all RARA rearrangements in APL.¹¹  Splicing of PRKAR1A exon 2 to RARA exon 3 generates a shorter out-of-frame fusion transcript possibly encoding a carboxy-truncated chimeric RIα protein.

PKA is a multimeric protein consisting of 2 catalytic subunits complexed with a regulatory subunit dimer.¹³  Binding of cAMP to the regulatory subunits produces a conformational change that causes dissociation and de-inhibition of the catalytic subunits.¹⁴  Type-I regulatory subunits are generally associated with cytosolic isoforms of PKA, whereas type-II regulatory subunits, capable of interacting with A kinase–anchoring proteins, are found in organelle-localized isoforms of PKA.¹³ 

PRKAR1A is involved in another gene rearrangement with the RET proto-oncogene in papillary thyroid carcinomas.^15,16  Inherited null mutations of PRKAR1A have been reported in Carney complex^17,18 PRKAR1A-inactivating mutations or down-regulation have also been found in some sporadic adrenocortical tumors.¹⁹  In our case, fusion of the RIα dimerization domain to RARα may be involved in dysregulated RARα homodimerization or heterodimerization with RXR and potentially deregulation of PKA through disruption of RIα.

Mouse models of APL have varying phenotypic features that depend on the RARA fusion partner.²⁰  It is conceivable that dysregulation of PKA by disruption of RIα may have contributed to this patient's atypical APL presentation. The ATRA responsiveness in this case is unknown because multiple agents were used. However, the remarkable clinical response suggests that the fusion gene products were sensitive to either As₂O₃ or ATRA or to both. An in vitro model of this variant could help clarify this issue, but based on current knowledge of the mechanisms of As₂O₃ and ATRA in APL, we propose that both agents may be effective. First, because the RARα component of the RIα-RARα fusion protein is identical to other ATRA responsive RARα fusion proteins and the truncated RIα component lacks nuclear corepressor interacting domains, it is conceivable that the RIα-RARα protein is responsive to pharmacologic doses of ATRA. Second, there may be synergic effects because of the involvement of PRKAR1A in the fusion. Gene expression profiling of the APL-derived NB4 cell line identified PRKAR1A as one of the genes that is up-regulated in response to ATRA treatment.²¹  Thus, ATRA treatment could also up-regulate expression of the in-frame PRKAR1A-RARA fusion transcript or the shorter out-of-frame fusion transcript, both of which might encode defective RIα proteins with dominant-negative effects on normal RIα. The resulting constitutive activation of PKA activity could lead to increased responsiveness to As₂O₃, which is known to have synergic effects with cAMP in APL.^22,23 

Although this report describes a single case expressing PRKAR1A-RARA, the nature of the cryptic cytogenetic lesion raises the possibility that other cases of variant APL carrying the same lesion may have been missed, even with RARA-specific FISH probes.

Arsenic trioxide was provided by Pharmalab for use in patients registered on the APML4 trial under the auspices of the Australasian Leukaemia and Lymphoma Group. The BAC probe (RP11–120M18) was kindly provided by Elizabeth Baker (King Edward Memorial Hospital, Perth, WA).

Contribution: A.C. designed and performed research, analyzed and interpreted data, and wrote the paper. M.A.D. provided clinical care; collected, analyzed, and interpreted data, and wrote the paper. K.S. performed research and analyzed data. S.O. analyzed and interpreted data and supervised data collection. A.S. provided clinical care and collected data. L.J.C. designed research, analyzed and interpreted data, and wrote the paper. H.I. analyzed and interpreted data and wrote the paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Alberto Catalano, Institute of Haematology, Royal Prince Alfred Hospital, Missenden Rd, Camperdown NSW 2050, Australia; e-mail: alberto.catalano@email.cs.nsw.gov.au.