APL is characterised by chromosomal rearrangements involving the gene encoding the myeloid transcription factor Retinoic Acid Receptor Alpha (RARα). While RARA is fused to PML in the majority of cases consequent upon the t(15;17)(q22;q21), significant insights can be gained into pathogenesis through analysis of APL involving alternative fusion partners. It is now apparent that the fusion partner has a critical bearing on disease biology, particularly with respect to sensitivity to molecularly targeted therapies i.e. all transretinoic acid (ATRA) and arsenic trioxide (ATO). While ATO sensitivity is restricted to PML-RARA+ APL, the fusion partner influences sensitivity to ATRA, with involvement of STAT5b or PLZF (ZBTB16) being associated with resistance and poorer clinical outcome. In the European Working Party study (Blood 2000;96:1297-1308) we previously established that ~10% of morphological APL cases lack the classic t(15;17). While PML is involved in the majority due to insertion events or more complex rearrangements, the molecular basis of disease in some patients has been unexplained. Analysis was undertaken in a normal karyotype APL (WBC 3.8 x 109/l, plt 7 x 109/l) presenting in a 37 year old man, PML-RARA negative by RT-PCR, with intact PML nuclear bodies on immunofluorescence. Due to the suggestive morphology, the patient was commenced on ATRA as a single agent (45mg/m2) pending further molecular investigation. ATRA induced maturation of the leukemic blasts, apparent in bone marrows taken at day 7 and day 16, associated with normalization of the platelet count (100 x 109/l by day 25). Diagnostic material was subject to gene expression profiling (TaqMan® Array Card, Life Technologies) and on unsupervised analysis clustered with PML-RARA+ cases, with a common signature associated with upregulation of HGF and FGF13 and downregulation of HOXA7, HOXA9 and MEIS1. 5’ RACE-PCR confirmed involvement of RARA and identified the fusion partner as IRF2BP2 (Interferon Regulatory Factor 2 Binding Protein 2), which is located at 1q42. The cDNA breakpoint involved fusion of exon 1 of IRF2BP2 to exon 3 of RARA leading to retention of the regions mediating binding to DNA, retinoic acid (ligand binding domain, LBD) and RXR. Chromosome 1 painting was undertaken, which appeared normal and a reciprocal RARA-IRF2BP2 fusion transcript was not detected, consistent with formation of IRF2BP2-RARA by cryptic insertion. Whole exome sequencing identified 17 non-synonymous exonic single nucleotide variants, three (with predicted functional impact) were in genes known to be recurrently mutated in AML (LRP1B, GATA2 and EZH2). The patient’s treatment response was tracked by real-time quantitative PCR (sensitivity 1 in 10e4), which showed persistent high level IRF2BP2-RARA fusion transcripts following single agent ATRA. The patient was commenced on the PETHEMA treatment schedule, achieved molecular remission (CRm) following the first course of chemotherapy and remains in ongoing CRm in the marrow 32 months from diagnosis.

IRF2BP2 is a downstream target of p53 and repressor of the cell cycle regulator NFAT1; it is fused to CDX1 as a result of the t(1;5)(q42;q32) in mesenchymal chondrosarcoma, has been reported to be mutated in CLL and primary CNS lymphoma, but has not previously been implicated in myeloid malignancy. To investigate the biology of IRF2BP2-RARα, co-immunoprecipitation assays were performed in 293T cells using tagged constructs, which revealed that the fusion protein has the capacity to self-associate. In reporter assays, IRF2BP2-RARα behaved similarly to PML-RARα, inducing repression at retinoid response elements, which was overcome by 1μM ATRA. Full length IRF2BP2-RARA was cloned into the MSCV vector, expressed in lin- murine stem/progenitor cells and replating assays performed in three conditions (SCF, IL3 + IL6/ SCF, IL3, IL6 + G-CSF/ SCF, IL3, IL6, G-CSF + GM-CSF). Expression of IRF2BP2-RARA consistently induced large numbers of third round colonies under all 3 conditions; whereas, colony formation was completely abrogated in the presence of 1μM ATRA in accordance with the sensitivity to retinoid observed in vivo.

In conclusion, these data extend the range of malignancies involving IRF2BP2, identifying a novel retinoid sensitive subtype of APL and provide further evidence that forced dimerization of RARα mediated by the fusion partner plays a critical role in leukemogenesis.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution