Next‐generation sequencing in two cases of de novo acute basophilic leukaemia

Abstract Acute basophilic leukaemia (ABL) is a rare subtype of acute myeloid leukaemia (AML); therefore, few data are available about its biology. Herein, we analysed two ABL patients using flow cytometry and next‐generation sequencing (NGS). Two cell populations were detected by flow cytometry in both patients. In Case no. 1, blasts (CD34+, CD203c−, CD117+, CD123dim+) and basophils (CD34−, CD203c+, CD117±, CD123+) were identified, both of which were found by NGS to harbour the 17p deletion and have loss of heterozygosity of TP53. In Case no. 2, blasts (CD33+, CD34+, CD123−) and basophils (CD33+, CD34+, CD123+) were identified. NGS detected NPM1 mutations in either blasts or basophils, and TET2 in both. These data suggest an overlap of the mutational landscape of ABL and AML, including TP53 and TET2 mutations. Moreover, additional mutations or epigenetic factors may contribute for the differentiation into basophilic blasts.


Abstract
Acute basophilic leukaemia (ABL) is a rare subtype of acute myeloid leukaemia (AML); therefore, few data are available about its biology. Herein, we analysed two ABL patients using flow cytometry and next-generation sequencing (NGS). Two cell populations were detected by flow cytometry in both patients. In Case no. 1, blasts (CD34 + , CD203c − , CD117 + , CD123dim + ) and basophils (CD34 − , CD203c + , CD117 ± , CD123 + ) were identified, both of which were found by NGS to harbour the 17p deletion and have loss of heterozygosity of TP53. In Case no. 2, blasts (CD33 + , CD34 + , CD123 − ) and basophils (CD33 + , CD34 + , CD123 + ) were identified. NGS detected NPM1 mutations in either blasts or basophils, and TET2 in both. These data suggest an overlap of the mutational landscape of ABL and AML, including TP53 and TET2 mutations. Moreover, additional mutations or epigenetic factors may contribute for the differentiation into basophilic blasts.

K E Y W O R D S
acute basophilic leukaemia, gemtuzumab ozogamicin, next-generation sequencing was diagnosed with pneumonia and treated with tazobactam/piperacillin (TAZ/PIPC) and teicoplanin at another hospital. Three weeks before presentation, bone marrow aspiration was performed for the evaluation of leukopenia, which revealed an increased number of blasts and immature basophils. Two weeks before presentation, he was diagnosed with acute myocardial infarction (AMI), for which emergent percutaneous coronary intervention was performed. After AMI treatment, he was referred to our hospital for ABL treatment.
Blood examination on admission to our hospital showed a leucocyte count of 0.52 × 10 9 /L (10% blasts and 6% immature basophils). Bone marrow aspiration revealed a further increase in blasts (38.2%) and immature basophils (33.8%; Figure 1A), but no dysplasia. Toluidine blue staining was positive for immature basophils ( Figure 1B). Electron microscopic examination demonstrated basophilic granules ( Figure 1C). Flow cytometry detected two populations, namely blasts (CD34+, CD203c-, CD117+, CD123dim+) and basophils (CD34-, CD203c+, CD117±, CD123+; Figure 1D). Blasts showed basophilic differentiation. Although basophils in the bone marrow were less than 40%, his condition was consistent with ABL, and prednisolone to prevent symptoms due to excessive release of histamine. 1 Induction therapy with DNR/AraC (daunorubicin 50 mg/m 2 on days 1-5 and cytarabine 100 mg/m 2 on days 1-7) was ineffective. Six weeks later, re-induction therapy with MEC (mitoxantrone 8 mg/m 2 , etoposide 100 mg/m 2 and cytarabine 1 g/m 2 ) was administered. However, it resulted in significant inflammation and temporal cognitive impairment, possibly due to tumour lysis syndrome on day 2. Therefore, it was discontinued. He was not tolerant to high-dose conventional chemotherapy. Four weeks later, gemtuzumab ozogamicin (GO; 10 mg/body) was administered, because the tumour cells express CD33. Following administration of GO, he developed an infusion reaction and was agitated again. He did not respond to any chemotherapeutic regimens and passed away 4 months after the initial diagnosis.

| Case no. 2
A 72-year-old Japanese man with a medical history of mitral stenosis and benign prostatic hyperplasia was referred to our institute for high-grade fever. Two weeks before presentation, he had fever and diarrhoea, which resolved simultaneously. On admission, he had high-grade fever and hypotension. He was admitted to our hospital for suspected urosepsis as the urinalysis showed bacteriuria.
For urosepsis treatment, he received meropenem and vancomycin. As hypotension persisted even after recovery from sepsis, he was administered vasopressor during induction chemotherapy.
We used histamine blockers and prednisolone similar to that in Case no. 1. After induction therapy with DNR/AraC (daunorubicin 40 mg/m 2 day 5-7 and cytarabine 80 mg/m 2 day 1-7), hypotension was resolved and haematological CR was achieved at day 33.
However, soon after, he had a relapse of the disease. He was refractory to subsequent chemotherapeutic regimens including mitoxantrone (MIT) monotherapy and MIT/AraC. He died 8 months after initial diagnosis.

| RE SULTS AND D ISCUSS I ON
In the present study, we presented two cases of ABL. We considered differential diagnosis of secondary cause of acute basophilic leukaemia such as blast phase of CML, AML with t(8;21) and basophilia, AML with t(6;9)(p23;q34), monocytic/monoblastic AML  . 2). A, Basophilic granules (black arrow) were present in the bone marrow (May-Giemsa stain, oil immersion lens, original magnification×1000). B, Flow cytometry analysis of the bone marrow: tumour cells were divided into two populations, blasts (CD33+, CD45dim+ and CD123-) and basophils (CD33+, CD45dim+ and CD123+). C, Nextgeneration sequencing data cells with 17p10-12 translocation led to loss of TP53, 7 we suspected that both cell types lacked TP53 activity. Moreover, either blasts or basophils had a point mutation in TP53 (R175H). In Case no. 2, two TET2 mutations at VAF of 40.9% and 5% and a NPM1 mutation at a VAF of 20.5% were identified. Both basophils and blasts seemed to carry TET2 mutations. However, we could not determine the population harbouring the NPM1 mutation.
Some molecular pathways may influence the phenotype and development of ABL. In four male infants with ABL, t(X;6)(p11;q23) and c-MYB-GATA1 mutations were reported. These were the first recurrent mutations of ABL 4 and promoted basophilic differentiation by induction of IL-33, NGF, IL1RL1 and NTRK1 expression. 8 c-Myb was stimulated by p16INK4a. 9 In Case #2, NPM1 through interaction with p16INK4a may influence c-Myb. 10 C/EBPα upregulation may be also associated with the development of ABL.
TET2 encodes an epigenetic modifier, which is known to act as an upstream regulator of mast cell and basophil lineage commitment.
The absence of TET2 causes up-regulation of C/EBPα (basophilspecific genes). 11 Therefore, we hypothesized that an additional NPM1 mutation, along with a TET2 mutation, could act synergistically to drive the leukemic transformation. 12 TET2 mutation combined with the epigenetic modification of C/EBPα may have modulated downstream molecules to induce basophil differentiation. As the relationship between TET2, NPM1, p53 and c-MYB-GATA1 has not been described yet, further studies are needed to reveal the molecular link.
In conclusion, we reported two ABL cases using NGS analysis.
Although mutations in these cases (TP53, NPM1 and TET2) are common in AML, they have not been previously reported in ABL. This suggests an overlap of the mutational profiles of ABL and AML. Case no. 1 had a complex karyotype with TP53 loss, whereas Case no.
2 had a normal karyotype with NPM1 and TET2 mutations, which suggests that epigenetic factors may promote differentiation into basophilic blasts. As cytogenetic and mutations of ABL are heterogeneous, further research is necessary for a better understanding on the factors that encourage basophilic differentiation in ABL.

ACK N OWLED G EM ENTS
The authors thank the patients and their families, all the medical staff members involved in treating this patient and Dr Ohno and Daiki Shimomura from Tenri Hospital for electron microscopy analysis (Case no. 1).

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest associated with the present study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.