When a patient presents with increased lymphoblast counts in either their peripheral blood or bone marrow aspirate and concurrently tests positive for the Philadelphia (Ph) chromosome, Bcell acute lymphoblastic leukemia (B-ALL) with BCR::ABL1 fusion and chronic myeloid leukemia (CML) in B-lymphoblastic crisis are among the malignant diseases that must be considered in the differential diagnosis. However, the differentiation of these malignancies based solely on morphological evaluation and flow cytometry is difficult, because the leukemic blast cells in both diseases can appear similar and share myeloid/lymphoid surface markers in flow cytometry [1]. In addition to considering the clinical presentation and progression of the disease, several laboratory protocols involving FISH can be used to aid in achieving a definitive diagnosis [1-3]. In this case report, we showcase a new variation of the FISH protocol that can be used to resolve the aforementioned ambiguities. To the best of our knowledge, this diagnostic protocol has yet to be reported in the literature. By performing separate FISH assays on two flow cytometrically sorted cell populations, we re-evaluated a B-ALL diagnosis to CML in Blymphoblastic crisis in the patient. This study was approved by the institutional review board of Yongin Severance Hospital, Yongin, Korea (IRB No. 9-2024-0010).

A 69-year-old male with dyspnea and general weakness was referred to Yongin Severance Hospital in September 2022. The complete blood count revealed leukocytosis, anemia, and thrombocytopenia (white blood cell count: 55.35×109/L; hemoglobin: 66 g/L; platelet count: 50×109/L). On initial evaluation, hepatosplenomegaly and lymphadenopathy were not observed. The blasts counts were increased in the peripheral blood (24% of all nucleated cells) without basophilia and in the bone marrow aspirate (87.6% of all nucleated cells). Flow cytometric analysis identified an abnormal blast population expressing CD34, cCD79a, CD19, CD38, CD7, HLA-DR, CD33, cCD22, and CD10. Karyotyping showed t(9;22)(q34;q11.2) without additional chromosome anomaly, and quantitative reverse transcription-nested PCR (qRTPCR) analysis indicated major BCR::ABL1 rearrangement (b2a2 subtype).

The patient was initially diagnosed with B-ALL with t(9;22) (q34;q11.2); BCR::ABL1 and was therefore prescribed induction chemotherapy (vincristine, daunorubicin, and prednisolone) and the tyrosine kinase inhibitor (TKI) imatinib mesylate. Follow-up bone marrow examination at 1-month post induction revealed a hypercellular marrow (70%) with myeloid hyperplasia and 4.6% blasts. The qRT-PCR-assayed BCR::ABL1 transcript level was 0.1% on the International Scale. However, the disease relapsed 11 months after the initial diagnosis, prompting its re-evaluation. A bone marrow examination was performed, followed by flow cytometric sorting of the blast cell and neutrophil (non-blast) populations guided by CD45 and side scatter patterns. Subsequent separate FISH analyses of the blast and non-blast cells confirmed the presence of the BCR::ABL1 fusion in both populations (Fig. 1).

Remarkably, 91.1% of 234 blast cells and 82.9% of 175 neutrophils exhibited the BCR::ABL1 fusion, solidifying the diagnosis of CML in B-lymphoblastic crisis. This case highlighted the enhanced diagnostic precision of this integrated approach of combining flow cytometric cell sorting and FISH analysis, which will be particularly valuable in scenarios with increased blast cell counts.

After the final diagnosis, the TKI was changed from imatinib to dasatinib, and hematopoietic stem cell transplantation (HSCT) was discussed in the event the patient proved refractory to the new drug regimen. Fortunately, treatment was effective and the patient showed negative minimal residual disease (as per qRTPCR- based BCR::ABL1 measurements) approximately 4 months after dasatinib treatment. The patient continues to visit the clinic as an outpatient and has not required HSCT. A timeline of the disease diagnoses, treatments, and monitoring for this patient is provided in Fig. 2.

Distinguishing between de novo B-ALL with BCR::ABL1 fusion and CML in B-lymphoblastic crisis is crucial for informing treatment decisions, as patients with the latter disease are typically advised to receive a brief course of TKI therapy together with chemotherapy followed by allogeneic HSCT, whereas patients with de novo B-ALL with BCR::ABL1 fusion who respond satisfactorily to chemotherapy and TKI are typically not considered for HSCT [4-7].

Most patients with CML are initially diagnosed in the chronic phase. Approximately 5% of patients are diagnosed in the accelerated phase/blast phase, and of those individuals, 20–30% are in the lymphoblastic crisis phase [8, 9]. In these patients, the presence of basophilia, myeloid hyperplasia, and micromegakaryocytes may be suggestive of underlying CML. However, CML in Blymphoblastic crisis can potentially be misdiagnosed as de novo B-ALL with BCR::ABL1 fusion without a documented history of CML [1]. Both diseases exhibit similar flow cytometric profiles, making a definitive diagnosis challenging. However, they can be distinguished according to several differences at the time of diagnosis [1, 2, 7, 9-14] (Table 1). The isoforms of BCR::ABL1 may aid diagnosis, since the p210 isoform is common in CML whereas the p190 isoform predominates in de novo B-ALL. The most reliable diagnostic method is to investigate the cell lineages that harbor the BCR::ABL1 translocation, given that CML in B-lymphoblastic crisis involves both myeloid and lymphoid cells since it is fundamentally a cancer originating from a myeloid progenitor [2]. Discrepancy between the percentage of blast cells in bone marrow aspirates and that of BCR::ABL1-positive cells in the interphase FISH test would indicate the presence of non-blast BCR::ABL1- positive clones, providing indirect evidence in support of CML [2]. Discordance between the percentage of blast cells and the size of BCR::ABL1 clones can also be supportive of CML [7].

Our patient was diagnosed with a CML in B-lymphoblastic crisis for several reasons. First, the p210 BCR::ABL1 fusion was detected, and it was not restricted to lymphoblasts but also presented in neutrophils at relapse. Second, after 1 month of induction, the bone marrow showed relatively high cellularity with myeloid hyperplasia and residual blasts, implying disease initiation from a myeloid lineage rather than a lymphoid one. In this case, we sorted bone marrow cells into blast and neutrophil populations using flow cytometry and then used the FISH assay to confirm CML. The CD45 biomarker and side scatter profile were used to distinguish the two populations, because the blast cell population displays a weak CD45 signal and low side scatter, allowing effective sorting. Several diagnostic methods for investigating BCR::ABL1 fusion in separate cell populations have been introduced in the past, such as identifying the nuclei of each cell type using 4′,6-diamidino-2-phenylindole (DAPI) staining or culturing each cell type to observe the occurrence of Ph translocation in neutrophils [2, 15]. However, differentiating cells by their nuclei requires visual inspection of many cells and can be challenging owing to artifacts, and cell culturing by lineage takes considerable time and effort. We expect our proposed method to be more convenient and definitive in cases where patients have greatly increased blast cell counts, where the relative scarcity of band/segmented neutrophils could make their visual localization using the DAPI-based FISH protocol challenging.

In a closely related context, the distinction between B-ALL with BCR::ABL1 fusion and CML-like features is increasingly being emphasized in the literature, and the level of discrepancy between post-treatment qRT-PCR results and molecular minimal residual disease measurements (e.g., immunoglobulin/T-cell receptor (Ig/TCR) rearrangement) has been suggested as a distinguishing feature [16, 17]. However, we were unable to incorporate this analysis for this patient since the approach requires Ig/TCR rearrangement testing at the time of diagnosis, which was not performed.

In our case, we sorted cells with flow cytometry using only the CD45 marker. This method is straightforward and we deemed it sufficient for distinguishing non-blast cells, especially neutrophils. However, this protocol has limitations in accurately identifying cell populations. Incorporating additional cell markers specific to blasts, such as CD19 and/or CD34, would likely enable more precise population grouping.

In conclusion, we report an alternative protocol whereby cell sorting and the FISH assay are combined for differentiating between B-ALL with BCR::ABL1 fusion and CML in B-lymphoblastic crisis. Further evaluation of this new integrated approach and analyses of its comparative performance against various diagnostic protocols will be needed to validate its effectiveness and efficiency.