Management of ETV6‐ABL1‐positive childhood acute lymphoblastic leukaemia: report of two cases, a literature review and a call for action

The 5-year overall survival (OS) of children with acute lymphoblastic leukaemia (ALL) has approached 90% by using risk-stratified chemotherapy intensive treatment protocols. The remaining 10% consists of ALL cases with rare genetic subtypes, which usually have primary resistance to standard chemotherapy regimens.1 Alternative treatment approaches are necessary for these rare subtypes of ALL. The ETS variant transcription factor 6-Abelson proto-oncogene 1, non-receptor tyrosine kinase [ETV6-ABL1, previously known as translocation-ETS-leukaemia-ABL (TEL-ABL)] was first observed in a child with B-cell precursor ALL (BCP-ALL),2 occurs in <0·2% of individuals aged <18 years, with only 16 cases in childhood ALL described.2-10 The ETV6-ABL1 rearrangement is complex, including at least three chromosomal breaks and the generation of two fusion products, both of which have elevated activity of tyrosine kinase (TK).10 Currently, ETV6-ABL1-positive ALL has been assigned to a heterogeneous high-risk subtype denominated Philadelphia chromosome (Ph)-like ALL. The management of Ph-like ALL has not been uniform.3 Similarly, the treatment approaches have not been consistent among children and adolescents with ETV6-ABL1-positive ALL, although there is strong evidence that haematopoietic neoplasms harbouring ETV6-ABL1 rearrangements are responsive to TK inhibitors (TKIs).11 In the present report, we describe the clinical and long-term outcomes of two children with ETV6-ABL1-positive ALL, review the literature and propose strategies to improve the management of this subtype of ALL. A previously healthy 7-year-old girl was admitted to our hospital with a 4-day history of fever. Laboratory evaluation demonstrated a white blood cell count (WBC) of 102.5 × 109/l, with 73% blasts on the differential. Bone marrow (BM) evaluation revealed a hypercellular marrow with 82% lymphoblasts by morphology. By flow cytometry, lymphoblasts expressed human leucocyte antigen-DR isotype (HLA-DR), CD123, CD19, CD10, CD20, CD58, CD79a and terminal deoxynucleotidyl transferase (TdT), and lacked T-cell expression myeloid and monocytic markers. Conventional cytogenetics analysis revealed a normal karyotype (46, XX). Fluorescence in situ hybridisation (FISH) analysis revealed negative results for breakpoint cluster region (BCR)-ABL1, histone-lysine N-methyltransferase 2A (KMT2A) and ETV6-Runt-related transcription factor 1 (RUNX1) rearrangements. Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis showed ETV6-ABL1 fusion transcripts. The patient was treated according to the Chinese Children's Leukemia Group (CCLG)-ALL-2008 protocol12 and attained complete remission (CR) by morphological examination of the BM on day 33. Measurement of minimal residual disease (MRD) was negative by both flow cytometry and PCR analysis. ETV6-ABL1 fusion transcripts were not detected by PCR as well. The leukaemia relapse was noted 32 months after diagnosis, with a large pelvic mass and BM involvement. The features of the leukaemia cells at diagnosis and relapse were similar, including the ETV6-ABL1 transcripts. The patient was treated with dexamethasone, vincristine, mitoxantrone and L-asparaginase. Imatinib at a daily dose of 300 mg/m2 was started on the first day of salvage therapy. The therapeutic response was rapid and profound, as demonstrated by the absence of the ETV6-ABL1 transcripts on day 7 of induction. However, cells with a leukaemia phenotype were detected by flow cytometry at a level of 2·01 × 10−4. The patient underwent allogeneic hematopoietic stem cell transplantation (HSCT). The post-transplant course was uneventful. By flow cytometry, MRD has been negative (<1 × 10−4). Imatinib was started 28 days after the HSCT and maintained for 2 years. The patient has remained in second CR for >3 years after HSCT. A previously healthy 3-year-old boy was admitted to our hospital in October 2014. Initial laboratory evaluation revealed a WBC count of 15·8 × 109/l, with 29% blasts on the differential. BM analysis showed the normal cellularity replaced by lymphoblasts expressing HLA-DR, CD123, CD19, CD10, CD58, CD34, CD79a and TdT. Cytogenetic and FISH analyses were unremarkable (46, XY, lacking BCR-ABL1, KMT2A and ETV6-RUNX1 rearrangements). RT-PCR analysis showed ETV6-ABL1 fusion transcripts. The patient received treatment according to the CCLG-ALL-2008 treatment protocol12 and a TKI was not administered. BM analysis on remission induction day 15 of induction showed no evidence of blasts by morphology. Flow cytometric analysis revealed 5.0 × 10−3 cells with the leukaemia phenotype. On day 33, there was no evidence of disease by morphology or flow cytometry (sensitivity <1 × 10−4). ETV6-ABL1 transcripts were undetectable. To date, he has been in the first CR for >5 years. Our present study illustrates the heterogeneity of ETV6-ABL1-positive paediatric ALL. The degree of the response of our first patient, including negative PCR for the ETV6-ABL1 transcripts on day 7 of salvage therapy, was unexpected and suggests that ETV6-ABL1-positive ALL is sensitive to TKIs. The second patient treated successfully with conventional chemotherapy indicates that prognostic implications of this genetic abnormality may vary. Of interest, our second case had a good response to standard chemotherapy and treated as a low-risk ALL. The clinical and outcome features of previously reported paediatric cases with this genetic abnormality also corroborate the variability of clinical and biological features, including 13 with progenitor B-cell ALL and two with T-cell ALL (Table 1). In adults, ETV6-ABL1-positive leukaemias have included ALL, acute myeloid leukaemia, atypical chronic myeloid leukaemia, chronic eosinophilic leukaemia, myelodysplastic and myeloproliferative neoplasms. Therefore, the morphological and immunophenotype distribution pattern of ETV6-ABL1-positive cases resemble those of classic BCR-ABL1-positive leukaemias. The rarity of the former might be due in part to the complexity of the rearrangement, which requires at least three break-and-join events.13 46,XY,inv(1),t(1;9),t(1;9;12) del(1),t(9;12),t(1;9) CNS irradiation re-induction CHT 47,XXY,del(6), ins(9;12),inv(9) Given the marked clinical and biological similarities among BCR-ABL1- and ETV6-ABL1-positive leukaemias and the impossibility of predicting reliably those patients who could be cured with conventional chemotherapy alone, we propose that all paediatric ETV6-ABL1-positive cases be treated with a TKI starting in induction remission and continuing during the entire treatment. This strategy has now been considered the standard of care for children with BCR-ABL1-positive ALL.14 Furthermore, we suggest that ETV6-ABL1-positive leukaemia be classified as a Ph+ variant instead of a Ph-like ALL. Finally, we recommend ETV6-ABL1 genetic analysis be included in the testing panel of children with newly or relapsed ALL and call for international ALL groups to collaborate to determine the true incidence of ETV6-ABL1-positive leukaemias among children and adolescents, to refine the treatment approaches for patients with this subtype of ALL. Shaoyan Hu and Jiajia Zheng designed the study; Peifang Xiao, Li Gao, Jing Ling, Qin Lu and Xuanxuan Shi provided patient material and clinical data; Raul Ribeiro, Shaoyan Hu and Jiajia Zheng analysed data. Jiajia Zheng, Shuiyan Wu and Yixin Hu wrote the manuscript; Raul Ribeiro and Shaoyan Hu reviewed the results, commented on draft manuscripts and revised the manuscript; finally, all the authors critically reviewed the manuscript and gave their approval. The authors declare that they have no conflict of interest. This work was supported by the following grants: 2020ZKPB02, BE2017659, CXTDA2017014, SZZX201504, SZS201615, SS201809 and National Clinical Research Center for Hematological Disorders. Raul C. Ribeiro, MD, is partially funded by National Cancer Institute (NCI) grant CA21765 and by the American Lebanese and Syrian Associated Charities (ALSAC). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.