BCR/ABL1-Like Acute Lymphoblastic Leukemia: From Diagnostic Approaches to Molecularly Targeted Therapy

B-cell acute lymphoblastic leukemia (B-ALL) is a malignant transformation and differentiation of a clonal population of B-lineage precursor cells within the bone marrow, blood, and extramedullary sites, and is characterized by a wide range of sentinel genetic alterations, including aneuploidy and chromosomal rearrangements [1]. The detailed mechanism of malignant transformation of B-cells remains unknown. The majority of cases carry mutations in genes encoding lymphoid transcription factors, tumor suppressors, cell cycle regulators, and epigenetic modifiers [2, 3]. The presence of cytogenetic and molecular aberrations is now the basis of the updated WHO B-ALL classification, dividing B-ALL into 2 main categories: B-ALL with recurrent genetic abnormalities and B-ALL not otherwise specified. Recurring chromosomal alterations are the distinctive feature of ALL and include aneuploidy, t(12;21) [ETV6/RUNX1], t(1;19) [TCF3/PBX1], t(5;14) [IL3/IGH], t(9;22) [BCR/ABL1], and the rearrangement of MLL [4]. The prevalence of sentinel genetic aberrations varies within age-groups, contributing to the decline of survivorship with an increase in age [5]. While ALL is the most frequent neoplasm in children with cure rates exceeding 85.0% [6], long-term prognosis for adults remains inferior. According to the Cancer and Leukemia Group B study, the 3-year overall survival rate varies between younger patients (30–59 years) and the elderly (>60 years), achieving approximately 38.0% and 12.0%, respectively [7]. Among adolescent and adult patients with ALL, the presence of favorable genetic features, including hyperdiploidy and ETV6/RUNX1, is less frequent than in children, while the prevalence of unfavorable genetic abnormalities (including BCR/ABL1) is higher [8]. Recently, a novel high-risk subtype of B-cell ALL has been identified with a similar gene expression profile to the BCR/ABL1-positive ALL. It was first reported in 2009 by Mullighan et al. [9] from the Children’s Oncology Group (COG) and St Jude Children’s Research Hospital (SJCRH) and by den Boer et al. [10] from Dutch Childhood Oncology Group (DCOG). The newly recognized B-ALL subtype is called BCR/ABL1-like or Philadelphia chromosome-like (Ph-like) ALL. It has been included as a provisional entity in the WHO classification [4]. The majority of BCR/ABL1-like ALL cases carry alterations in key transcription factors involved in the B-cell development, including IKAROS family zinc finger 1, transcription factor 3 (E2A), early B-cell factor 1, and paired box 5 [11-13]. The incidence of BCR/ABL1-like ALL increases with age, accounting for 10.0–15.0% in children, 21.0% in adolescents, 27.0–27.9% in young adults (21–39 years), 20.4% in adults (40–59 years), and 24.0% in older adults (>60 years) [11, 14, 15]. BCR/ABL1-like subtype is associated with inferior outcomes, compared to other B-cell precursor ALL (BCP-ALL): BCR/ABL1-negative and non-BCR/ABL1-like ALL. Patients with this subtype have a higher risk of relapse, lower 5-year event-free survival rates, and overall survival rates. BCR/ABL1-like ALL is also associated with a high rate of nonresponse to first-line treatment, high relapse rate, and persistent minimal residual disease (MRD), compared to other BCP-ALL [14, 16-18]. BCR/ABL1-like ALL patients have higher WBC counts at the moment of diagnosis in comparison with other BCP-ALL patients [14].

Genetic alteration-activating kinase and cytokine receptor signaling are the hallmark of BCR/ABL1-like ALL [11, 16]. Kinase-activating genetic abnormalities can be categorized into several classes. These include rearrangements of cytokine receptor-like factor 2 (CRLF2) gene, V-abl Abelson murine leukemia (ABL) viral oncogene homolog gene-class fusions (ABL1, ABL2, CSF1R, PDGFRB, PDGFRA), Janus kinase 2 (JAK2) gene, erythropoietin receptor (EPOR) gene rearrangements, JAK/STAT-activating aberrations, Ras pathway mutations (KRAS, NRAS, NF1, PTPN11), and other uncommon fusions (NTRK3, PTK2B, BLNK, FLT3) [11, 14, 19, 20]. The rearrangements of CRLF2 remain the most frequent alteration in BCR/ABL1-like ALL. Approximately, 35.7–75.0% of BCR/ABL1-like ALL patients have CRLF2 overexpression [18, 21, 22]. CRLF2 encodes the thymic stromal lymphopoietin receptor (TSLPR). It heterodimerizes with interleukin-7 receptor alpha chain and leads to induction of JAK/STAT, Ras/MAPK, and PI3K/Akt/mammalian target of rapamycin (mTOR) pathway signaling and lymphoid development. CRLF2 is located in the pseudoautosomal region 1 on the Xp22.3 and Yp11.3 chromosomes [23]. Overexpression of CRLF2 occurs as a consequence of either translocation juxtaposing CRLF2 to the immunoglobulin heavy chain locus or an intragenic deletion of the pseudoautosomal region 1 of chromosomes X and Y (P2RY8-CRLF2) [23, 24]. Less frequently, activating point mutations (most often p.Phe232Cys) are present [25]. Although the rearrangements of CRLF2 are the most common alteration in BCR/ABL1-like ALL, approximately 5.0–10.0% of CRLF2-rearranged ALL cases are not BCR/ABL1-like ALL, since the gene expression signature lacks genetic alteration-activating tyrosine kinase signaling [26]. What is more, the diagnosis of BCR/ABL1-like ALL is challenging due to a possible concurrence of CRLF2 and BCR/ABL1 rearrangements [27]. According to Meyer et al. [28], BCR/ABL1-like ALL blasts with CRLF2 rearrangements demonstrate reduced sensitivity to glucocorticoids in vitro, which could be partially responsible for poor therapy outcomes. The overexpression of CRLF2 is associated with unsatisfactory response to chemotherapy and poor outcomes even in non-BCR/ABL1-like cases [22, 29]. What is more, among BCR/ABL1-like ALL patients, CRLF2-rearranged cases carry the worst prognosis [18]. It was confirmed that approximately 50.0% of these cases harbor mutations in the JAK family genes, most frequently in JAK2, rarely in JAK1 and JAK3 [11, 22, 25]. Activating point mutations within JAK2 occur in different locations than V617F. The most recurrent alteration is R683G [13, 30]. The concomitance of CRLF2 overexpression and JAK2 mutations is associated with inferior outcomes [11, 17]. Less frequently, gain-of-function somatic mutations (c.642A>T, c.814ins13delA, c.819ins12, c.820ins21, c.828ins7delT) within the interleukin-7 receptor alpha chain have been observed in the CRLF2-rearranged leukemias [31]. Several preclinical studies have shown that CRLF2-rearranged leukemias and leukemias with other JAK pathway alterations are susceptible to JAK inhibitors or a combination of JAK/STAT and mTOR pathway inhibitors [32-34]. Furthermore, overexpression of TSLPR on blast surface may be an immunotherapeutic target for chimeric antigen receptor T-cells [35, 36]. According to Qin et al. [35], TSLPR may play a significant role in the survival of leukemic cells, since late relapses occur with a retained expression of this cytokine receptor. This finding is in accordance with comparable CRLF2 expression frequency among patients with de novo and relapsed ALL observed by Konoplev et al. [37]. The second most common kinase pathway alterations within BCR/ABL1-like ALL are ABL-class fusions, present in approximately 10.0–19.0% of ALL cases, more frequently in children than in adults [11, 19, 20, 26, 38]. According to several studies, patients with ABL-class rearrangements may benefit from the addition of tyrosine kinase inhibitor (TKI) – dasatinib to chemotherapy [16, 39, 40]. The rearrangements of JAK2 and erythropoietin receptor are other commonly observed alterations within BCR/ABL1-ALL and occur in approximately 6.5–27.3% and 4.0–6.3% of patients, respectively [14, 20, 41]. Interestingly, the JAK2 rearrangements are more frequently observed among young adults than children [14].

Flow cytometry (FC) can facilitate the identification of BCR/ABL1-like ALL. Overexpression of TSLPR on blast surface is predictive of CRLF2-rearranged ALL and can be identified by FC [42] shown in Figure 1. According to Ohki et al. [43], BCR/ABL1-like ALL cases have similar immunophenotype to BCR/ABL1-positive ALL. A relatively low expression frequency of aberrant myeloid antigen appears – CD27 and CD66c reaching in 44.0% and 36.4% of cases, respectively [43]. At the same time, Ohki et al. [43] observed relatively high CD45RA and CD99 expression in BCR/ABL1-like ALL in 80.0% and 90.9% of cases, respectively. However, the immunophenotypic signature distinctive for BCR/ABL1-like ALL has not been established so far [44].

Establishing BCR/ABL1-like diagnostic algorithm remains of clinical significance in view of prognostic relevance and the advent of clinical trials of targeted therapy. Unfortunately, the recognition of BCR/ABL1-like ALL is a major challenge due to the diversity of underlying genetic alterations. There are 2 ways to facilitate the identification of BCR/ABL1-like ALL: stepwise algorithms, which are economical, but require long turnaround time, or broad-based testing, which requires higher cost, but enables rapid recognition of potentially targetable alterations [45]. So far, several stepwise algorithms have been proposed to facilitate the detection of BCR/ABL1-like ALL. These include tiered algorithms and targeted fusion/gene panel testing. The targeted approach identifies key fusions associated with BCR/ABL1-like ALL. The tiered algorithm allows detecting BCR/ABL1-like ALL based on the quantification of selected predictor gene expression, using a low-density array (LDA) or quantitative real-time polymerase chain reaction (qRT-PCR) followed by sequential genomic profiling [18, 41, 46]. LDA with an array of 15 genes was developed for the first time by Harvey et al. [47], followed by the creation of other predictors with fewer genes by Heatley et al. [48] and Roberts et al. [49]. Chiaretti et al. [50] built a predictor based on qRT-PCR to assess the expression values of ten types of transcripts. Tiered algorithm with TaqMan-based LDA is currently routinely applied by the Children Oncology Group in newly diagnosed high-risk childhood B-ALL [51]. Patients identified positively for BCR/ABL1-like ALL gene expression profile undergo subsequent genetic testing to identify associated genetic alterations. Those with high CRLF2 expression eventually receive an analysis for the presence of CRLF2 rearrangements and JAK1/JAK2 mutations. Patients lacking CRLF2 overexpression undergo multiplex RT-PCR to detect other kinase fusions with subsequent transcriptome sequencing [52]. The St Jude’s Children Hospital adopted a different approach, with RNA-sequencing performed within the first 2 weeks of the remission induction. This approach allows early identification of targetable genetic abnormalities and implementation of TKIs [19]. Regardless of the incorporated algorithm, the main goal of the diagnostic analysis is to detect the underlying genetic alterations, since they are critical for the diagnosis, prognosis, and targeted therapy [51]. The proposed algorithms are listed in Table 1. Some recommend performing fluorescent in-situ hybridization (FISH) analysis using CRLF2 probe in subjects with high expression of CRLF2 in FC to identify BCR/ABL1-like ALL patients. The possible result of the analysis is shown in Figure 2.

The determination of an optimal therapeutic strategy in BCR/ABL1-like ALL is desired, considering poor outcomes in this entity with conventional treatment. Studies by Chiaretti et al. [50] revealed that BCR/ABL1-like ALL patients have a significantly lower complete remission (CR) rate and a higher risk of MRD persistence relative to non-BCR/ABL1-like ALL patients [21]. Furthermore, BCR/ABL1-like patients experience inadequate event-free survival and disease-free survival. In this study, MRD persistence implicated the performance of hematopoietic stem cell transplant (HSCT) in first CR (CR1); hence, significantly more BCR/ABL1-like patients underwent HSCT than non-BCR/ABL1-like patients. On the other hand, since the benefit of early MRD clearance in BCR/ABL1-like ALL patients remains questioned, the researchers consider post-induction therapy intensification, regardless of the MRD response status [56]. It remains debatable whether all adult BCR/ABL1-like ALL patients should receive an allogeneic HSCT in the first CR, irrespective of other indications [57]. So far, no recommendations on the use of allogeneic HSCT have been established in BCR/ABL1-like ALL patients in the CR1 [58]. According to El Fakih et al. [58], this modality should be recommended in CR1, with several factors considered during the evaluation of the candidates. These factors include the MRD status, presence of CRLF2, IKAROS family zinc finger 1, and JAK2 rearrangements, and availability of modern salvage therapies. Studies by Cho et al. [59] indicate that allogenic HSCT in BCR/ABL1-like ALL patients contributes to improved outcomes. The role of post-HSCT maintenance therapy with TKI in MRD-persistent patients remains debatable. It is questioned whether this approach may eradicate MRD and reduce relapse risk. Aldoss et al. [57] presented a case of a favorable outcome in a patient who received single-agent dasatinib after allogeneic HSCT. Moreover, in view of underlying molecular changes in BCR/ABL1-like ALL, the incorporation of TKIs into the therapy regimen is of considerable interest. The results of several preclinical studies signalized the activity of ABL-class inhibitors and JAK inhibitors in ABL-class mutant and JAK pathway mutant BCR/ABL1-like ALL, respectively [11, 16, 34, 60, 61]. Jain et al. [62] observed partial downregulation of JAK-STAT signaling in relapsed/refractory CRLF2-rearranged patients who received ruxolitinib in combination with chemotherapy. Studies of Böhm et al. [63] proved that the addition of ruxolitinib to induction therapy with vincristine, dexamethasone, and L-asparaginase increased the in vivo effect in the CRLF2-rearranged BCR/ABL1-like ALL xenografts. Promising outcomes of TKIs utilization were also reported by Roberts et al. [64]. They demonstrated the cytostatic effect of dasatinib and ruxolitinib monotherapy in BCR/ABL1-like ALL harboring ABL-class fusions and JAK-STAT-activating aberrations, respectively. Furthermore, a combination with dexamethasone enhanced this effect [64]. Since single-agent therapy with TKIs in BCR/ABL1-like ALL results in limited efficacy, a combination with other agents is under investigation. Studies by Gotesman et al. [40] and Tasian et al. [60] demonstrated that the addition of a mTOR inhibitor enhanced the antileukemic effect of TKIs. Single case reports showed clinical efficacy of a combination of ruxolitinib with chemotherapy in patients with chemoresistant BCR/ABL1-like ALL [32, 65]. On the other hand, a recently published case report by Byram et al. [66] showed successful management of a child with relapsed BCR/ABL1-like ALL with high-dose ruxolitinib in monotherapy, followed by haploidentical HSCT. There are also several anecdotal reports revealing encouraging results of dasatinib in BCR/ABL1-like ALL [39, 67-70]. Currently, several proceeding clinical trials are evaluating the safety and potency of the addition of dasatinib or ruxolitinib to standard chemotherapy in patients with BCR/ABL1-like ALL. Ongoing clinical studies are listed in Table 2. A synergistic effect of TKIs with B-cell lymphoma 2 inhibitors (e.g., venetoclax and navitoclax) was reported by Roberts et al. [64]. Both drugs promote apoptosis and demonstrate antitumor activity in ALL models in the preclinical studies. The phase I study using venetoclax and navitoclax in combination with chemotherapy in relapsed/refractory ALL patients revealed promising outcomes within all genomic subtypes, although in kinase-driven subtypes (BCR/ABL1-like and BCR/ABL1-positive ALL), responses were fewer [71]. Recently presented results of Aldoss et al. [72] demonstrated promising outcomes of novel therapies for relapsed/refractory BCR/ABL1-like ALL, including blinatumomab (a Bi-Specific Anti-CD19/CD3 BiTE® Antibody), inotuzumab ozogamicin (humanized monoclonal anti-CD22 antibody), chimeric antigen receptor T-cell therapy, and venetoclax-navitoclax anti-B-cell lymphoma 2 therapy [72]. According to the authors’ opinion, these therapies could improve the BCR/ABL1-like ALL treatment results by reducing disease chemoresistance.

BCR/ABL1-like ALL is characterized by the presence of genetic alteration-activating kinase and cytokine receptor signaling. Underlying molecular changes are complex and often undetectable with standard diagnostic techniques; therefore, identification of this disease subtype remains a clinical challenge. Considering that kinase fusions in BCR/ABL1-like ALL contribute to leukemogenesis, and given the inferior outcomes of this entity with conventional chemotherapy, there is a need to establish an optimal diagnostic and therapeutic approach. Since several clinical studies are currently investigating the addition of molecularly targeted agents to chemotherapy in BCR/ABL1-like ALL, the prognosis may improve, similarly to the progress observed in BCR/ABL1-positive ALL [73]. Given the fact that results of chemotherapy in BCR/ABL1-like ALL patients remain poor, it is believed that BCR/ABL1-like ALL patients may benefit from the participation in clinical trials evaluating the efficacy of molecularly targeted therapy based on detailed FC and molecular disease characteristics [60].

We would like to thank Kinga Gwóźdź, MSc, for performing blood and bone marrow smear, Jolanta Kiernicka-Parulska, MSc, and Anna Mierzwa, MSc, for the FC data analysis, and Anna Przybyłowicz-Chalecka, MSc, for the FISH study result documentation.

The authors have no conflicts of interest to declare.

The authors have no funding sources to declare.

Both authors, Anna Płotka and Krzysztof Lewandowski, have contributed significantly to the conception of the manuscript and reviewed the available literature. The manuscript was drafted by Anna Płotka and revised critically by Krzysztof Lewandowski. Both authors approved the final version, and both are accountable for all aspects of the work.