A novel dasatinib-sensitive RCSD1-ABL1 fusion transcript in chemotherapy-refractory adult pre-B lymphoblastic leukemia with t(1;9)(q24;q34)

Dasatinib, a wide-spectrum tyrosine kinase inhibitor (TKI), has shown notable clinical activity in Philadelphia chromosome positive (Ph) acute lymphoblastic leukemia (ALL) patients carrying a 9;22 translocation and the resulting BCR-ABL1 fusion gene.1 In other subtypes of ALL, its clinical efficacy and putative molecular targets are unknown. In this report we describe a pre-B ALL patient, who had chemotherapy resistant, but dasatinib sensitive disease and a t(1;9)(q24;q34) translocation. Sequencing revealed a novel RCSD1-ABL1 fusion transcript suggesting involvement of this fusion tyrosine kinase in leukemogenesis and sensitivity to dasatinib.

A 40-year old male was diagnosed with a pre-B ALL based on immunophenotyping (CD19, CD22, CD10, cCD79a , nTdT, CD34, CD38, HLA-DR). The bone marrow (BM) aspirate showed 80% blast cells. The cytogenetic analysis revealed a t(1;9)(q24;q34), described in more detail below. The patient received a modified CVAD induction therapy. After 2 induction courses, the BM had residual disease with 50% of leukemic blasts. At day 10 from the start of the 3 induction regimen, dasatinib was introduced at a dose of 140 mg once daily. Two weeks later morphological remission was confirmed, and only minimal residual disease (1.5% of leukemic cells) was detectable by FISH. The first consolidation therapy consisted of dasatinib 140 mg once daily and high-dose methotrexate. BM at four weeks post consolidation showed complete morphological and cytogenetic remission (CCyR) (1,020 cells analyzed by FISH). Dasatinib was continued until an allogeneic hematopoietic stem cell transplantation (HSCT) was performed four months after diagnosis. After transplantation, no TKI therapy was used.

At 12 months from HSCT, the patient experienced a cytogenetic relapse and a month later a hematologic relapse. Immunosuppressive drugs were discontinued, and the patient received a re-induction therapy with dasatinib 70 mg twice daily combined with one cycle of high-dose chemotherapy. The patient successfully reentered in CCyR, which has been maintained with dasatinib monotherapy for more than nine months.

Figure 1.Cytogenetic characterization of the t(1;9)(q24;q34) translocation. Detailed characterization of the translocation was made using (A) G-banding method, (B) multicolor FISH (Metasystems GmbH, Altlussheim, Germany) and (C, D) two different FISH probes. In panel (C) red signals show hybridization of 1p36 specific probe and green signals binding of 1q25 specific probe with chromosome 9 painting. In panel (D), BCR-ABL1 ES dual color extra signal translocation probe (Vysis®, Abbott Laboratories, IL, USA) was used showing two normal BCR signals from chromosome 22, but three ABL1 signals. No BCR-ABL1 fusion signal was detected either with FISH or PCR analysis with different primers for various BCR-ABL1 transcripts.

At the time of diagnosis, a clonal karyotype, 46,XY,t(1;9)(q24;q34)[5]/46,XY[3], was detected in 60% of BM cells (Figure 1A). No other clonal abnormalities were detected. The reciprocal nature of the t(1;9) aberration was confirmed by multicolor FISH (Figure 1B) and refined with FISH probes specific for chromosome 9 and 1p36 and 1q25 regions (Figure 1C). FISH for BCR-ABL1 gene fusion showed 2 normal BCR signals from chromosome 22, but a split signal of the ABL1 probe (Figure 1D), which referred to a translocation between ABL1 and an unknown gene located in chromosome 1 centromeric to 1q25 region.

Figure 2.RCSD1-ABL1 fusion gene resulting from a t(1;9)(q24;q34) translocation. Panel (A) shows an agarose gel of PCR products carried out with different RCSD1-ABL1 primer sets (a–e) (primer pairs aF + bR; aF + cR; bF + aR; bF + bR; bF + cR, respectively). Primers for RCSD1 were: forward; RCSD1-aF 5’-CCT-GAAGGACATGGAGGAAA-3’; RCSD1-bF 5’-CAGAGACCAATGCCAATGTG-3’; binding to exons 1–2. Primers for ABL1 were: reverse; ABL1-aR 5’-CTGCACCAGGTTAGGGTGTT-3’, ABL1-bR 5’-TGGTGTCCTCCTTCAAGGTC-3’ and ABL1-cR 5’- TCTGAGTGGCCATGTACAGC-3’; binding to exons 4–6. Lanes 1 and 14: DNA ladder; lanes 2, 5, 8, 10 and 12: PCR products of the patient cDNA; lanes 3 and 6: negative control cDNA; lanes 4, 7, 9, 11 and 13: no template control. Two PCR products were visible corresponding to sequencing results shown in panel (B). In the longer PCR product the fusion gene consisted of the three first exons of RCSD1 gene and in the shorter product lower band only the two first exons were included. In both cases, the RCSD1 fused to the exon 4 of the ABL1 gene. The same fusion transcripts were detected on lanes 2 and 12 as in lanes 8 and 10 but the alternatively spliced forms were less efficiently amplified. No PCR products were observed in control reactions with normal cDNA as a template.

Based on a previous publication by De Braekeleer et al.,2 involvement of the RCSD1 gene located in the long arm of chromosome 1 was suspected. Primers covering the kinase domain of ABL1 and most of the exons of RCSD1 were designed. PCR analysis from leukemic cells showed 2 positive products of slightly different molecular weights (Figure 2A). Sequencing revealed that the longer PCR product consisted of the first 3 exons of RCSD1 fused to ABL1 gene starting from exon 4 extending to the kinase domain coding region (Figure 2B). The shorter PCR product consisted of the first 2 exons of RCSD1 fused similarly to exon 4 of ABL1 (Figure 2B). These two different fusion transcripts were likely caused by alternative splicing of the fusion gene. The predicted oncogenic product of RCSD1-ABL1 is in-frame and encodes the tyrosine kinase domain of ABL.

Most previously described fusion genes involving ABL1 (BCR, ETV6, EML1, NUP214) fuse with exon 2 of ABL1.3–5 However, similar to RCSD1-ABL1, a novel fusion partner (SFPQ) was recently reported to fuse to exon 4 of ABL1 in a pre-B-ALL patient with a t(1;9)(p34;q34) translocation. 6 The fusion protein resulting from ABL1 exon 4 fusion lacks the amino acids W127-K183 to form intact SH2 domain.6 Murine experiments have shown that BCR-ABL constructs with inactivated SH2 domain are able to induce B-ALL-like disease,7 which further supports the central role of RCSD1-ABL1 in the development of leukemia in our patient. However, additional mechanisms may also exist, as the RCSD1 gene encodes a protein kinase substrate CapZIP which interacts with the actin capping protein CapZ. This may influence the cytoskeleton regulation and/or migration of the cells.8

In PhALL dasatinib is a promising novel treatment option.1 Our case report describes that TKIs may have clinical efficacy in other types of B-ALL as well. Furthermore, the clinical activity of dasatinib was recently described in a T-ALL patient carrying a NUP214-ABL1 fusion.9 Although these fusion genes may occur rarely, the therapeutic implications warrant that they be searched for in acute leukemia. The screen for such uncommon, but clinically significant fusion transcripts should be included in multiplex PCR panels used in the routine diagnostic work-up. Our data further underline the importance of dividing hematologic malignancies into molecularly defined and clinically meaningful disease entities, allowing for a rational and effective selection of optimal treatment modalities for each patient.10

Acknowledgments

the authors would like to thank the personnel at the Department of Pathology, Laboratory of Molecular Genetics, Helsinki University Central Hospital and Tampere University Hospital for their help with patient sample analysis.

Footnotes

• Funding: this work was supported by the Finnish special governmental subsidy for health sciences, research and training, by the Finnish Cancer Societies, Emil Aaltonen Foundation, Academy of Finland, Finnish Medical Foundation, Biomedicum Helsinki Foundation and KA Johansson foundation.