Two episodes of Ph+ acute leukemia with divergent Ig/TCR rearrangements in two patients with persistent BCR::ABL1 positivity: a 17-year follow-up

Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph⁺ ALL) remains a biologically and clinically heterogeneous disease. We wish to contribute to the ongoing discussion, in line with previously published findings,¹ by reporting two pediatric patients with Ph^⁺ ALL with persistent BCR::ABL1 positivity despite sustained minimal residual disease (MRD) negativity by immunoglobulin and T-cell receptor (Ig/TCR)-based assays. Both experienced a second disease episode after a long interval harboring different Ig/TCR targets.

The Philadelphia chromosome translocation (t(9;22)) results in the molecular juxtaposition of two genes, BCR (breakpoint cluster region) and the proto-oncogene ABL1, to form an aberrant BCR::ABL1 fusion gene on the derivative chromosome 22. This fusion is the hallmark genetic alteration of chronic myeloid leukemia (CML) but it is also detected in approximately 2-5% of pediatric and 25-30% of adult cases of acute lymphoblastic leukemia (ALL), and classified as Ph^⁺ ALL.² Whereas the breaks in ABL1 occur in the same region, two different breakpoint cluster regions affect the BCR gene. More than 90% of children with Ph⁺ ALL show a break in the ‘minor’ breakpoint cluster region between the BCR exons 1 and 2 resulting in a fusion protein of 190 kDa (p190).³ In contrast, the rearrangement of ABL with the ‘major’ breakpoint cluster region within BCR (p210) is the usual finding in CML⁴ and an occasional finding in Ph⁺ ALL.³ Ph^⁺ ALL is almost exclusively of B lineage, with less than 2% of cases of T lineage leukemia according to the pediatric Children’s Oncology Group (COG) and the European group (EsPhALL).^5-7 In contrast to adults, CML in pediatric patients is rare (1-2.2 cases per million per year, with 7-10% being diagnosed with blast phase).⁸ Children with CML in blast phase also present with a distinct predominance of the lymphoid phenotype (70-80% vs. 20-30% in adults) and a varying landscape of additional chromosomal aberrations.⁹ Those leukemic features complicate the distinction between pediatric de novo CML in lymphoblastic blast phase (CML-BP) and Ph^⁺ ALL. In recent years it became clear that also Ph^⁺ ALL cases are more heterogeneous than previously thought and that distinct subtypes, correlate with multilineage versus lymphoid-only BCR::ABL1 involvement and differ in outcome.¹⁰ This is also reflected in the recent consensus classification of myeloid neoplasms and acute leukemias with the term Ph^⁺ ALL with multilineage involvement.¹¹ In 20-30% of children (and more than 30% of adults^10,12) diagnosed with Ph^⁺ ALL the BCR::ABL1 fusion is present in a wider clone, involving myeloid cells, non-ALL B cells and T cells and termed CML-like ALL.¹³ The overall prognosis of CML-like ALL is similar to Ph^⁺ ALL, but whereas relapses are more common in typical Ph^⁺ ALL, early deaths from treatment toxicity are more common in CML-like ALL, highlighting the need for early distinction and tailored therapy.¹³ Additionally, diagnosing underlying CML may be relevant given the fact that hematopoietic stem cell transplantation (HSCT) remains the only curative option for CML-BP.¹¹

Ph⁺ALL is commonly monitored using two main polymerase chain reaction (PCR)-based targets: clonal (Ig/TCR) gene rearrangements, and the quantification of BCR::ABL1 levels, either at DNA or RNA level.¹⁴ This may complicate the decision on HSCT indication, as also outlined in a recent article by Short and colleagues if Ig/TCR MRD becomes negative while the BCR::ABL1 fusion transcript remains detectable.^15,16 Of note, Short and colleagues here refer to molecular MRD using next-generation sequencing-based technologies and not conventional Ig/TCR PCR MRD measurements.

As these discrepancies occur frequently, the COG and EsPhALL consortia only consider MRD Ig/TCR gene rearrangements for risk stratification as long as informative results for Ig/TCR MRD are available.⁵

Here, we report two pediatric patients with Ph⁺ ALL who, despite achieving repeated negative MRD by Ig/TCR-based assays, showed persistent BCR::ABL1 positivity, with a second episode of Ph⁺ ALL, harboring different Ig/TCR rearrangements compared to the first episode after long time intervals. Both cases demonstrate the importance of long-term observation and raise additional questions about optimal management for these patients. This study has been approved by the local ethics committee and informed consent was obtained from both patients.

The first patient is a boy of 20 years, who was initially diagnosed at the age of 2 years with a p190 BCR::ABL1-positive pre-B ALL, and was treated on the EsPhALL-2004 protocol, and randomized to the arm without imatinib. The Ig/TCR-based MRD declined rapidly to <0.01% by the end of induction (EOI), whereas RNA-based quantitative PCR (qPCR) BCR::ABL1 transcripts persisted at 1-4% throughout therapy, yet the child completed treatment in continuous complete remission and no further MRD follow-up was done. Eleven years later, at the age of 13 years, he developed a new episode of p190 BCR::ABL1-positive leukemia that demonstrated the same BCR::ABL1 fusion at DNA level but a different Ig/TCR rearrangement, favoring the interpretation of a second malignancy rather than a classic late relapse. He was retreated according to the EsPhALL2010 protocol with initially imatinib, later dasatinib (Figure 1) which yielded negativity by Ig/TCR MRD at EOI, but DNA-based qPCR BCR::ABL1 MRD remained persistently detectable at levels of 2-11%. HSCT at relapse was considered but because the patient did not have a suitable donor and given the rapid response it was decided not to consolidate with HSCT. Notably, discontinuation of therapy after maintenance did not provoke any expansion of the BCR::ABL1-positive cells. Tyrosine kinase inhibitor (TKI) monotherapy for over 4 years, including dasatinib, ponatinib, and asciminib, failed to eradicate the residual BCR::ABL1 positivity (Figure 1), using a DNA-based qPCR that was developed in the meantime. He has currently been without any therapy for over 1 year. More detailed analyses of marrow subsets revealed that BCR::ABL1 positivity involved a small population of T and B lymphocytes, with no evidence of myeloid involvement. This patient remains in complete remission more than 4 years after the second course of therapy, with negative Ig/ TCR PCR (Table 1; Figure 1).

At the age of 10 years, the second patient - a now 19-year-old man - was diagnosed with p190 BCR::ABL1-positive, hyperdiploid pre-B ALL. Ig/TCR MRD negativity was obtained at the end of consolidation (EOC) with the EsPhALL-2010 protocol that included imatinib. Six years after completing treatment he developed a new episode of p190 BCR::ABL1-positive leukemia, harboring the identical BCR::ABL1 fusion at DNA level but demonstrated, similar to the first patient, different Ig/TCR rearrangements, also favoring the interpretation of a second malignancy. He received only mild chemotherapy followed by chimeric antigen receptor T cells (CAR T), also because of a cardiomyopathy after first line chemotherapy, which resulted in swift clearance of Ig/TCR MRD on day 28 after CAR T. Although he lost CAR T cells soon after infusion, he has remained in continuous complete remission for the next 2 and a half years. MRD for BCR::ABL1 using DNA-based qPCR remained at approximately 3%. Despite sequential therapies with imatinib, dasatinib, and asciminib, the BCR::ABL1 MRD levels using a DNA-based qPCR have remained stable in the range of 0.2-1%, also without treatment now for over 1 year. Immunophenotypic sorting of the remission bone marrow showed BCR::ABL1 positivity in about 2% of T cells and 2% of normal B cells, while myeloid cells were negative. Ig/TCR MRD PCR in the subpopulations has not been performed (Table 1; Figure 1).

Figure 1.Swimmer plot of treatment history and molecular responses in the two patients. ALL: acute lymphoblastic leukemia; BCR::ABL1: Breakpoint Cluster Region-Abelson Tyrosine Kinase 1 fusion gene; EsPhAL: European intergroup study protocol for Philadelphia chromosome-positive ALL; TKI: tyrosine kinase inhibitor; Ig/TCR MRD: immunoglobulin/T-cell receptor minimal residual disease; MRD: minimal residual disease; CAR T: chimeric antigen receptor T-cell therapy; Off Tx: off treatment; Dx: diagnosis; neg: negative; Pos: positive; Imatinib: first-generation tyrosine kinase inhibitor; Dasatinib: second-generation tyrosine kinase inhibitor; Ponatini: third-generation tyrosine kinase inhibitor effective against T315I mutations; Asciminib: allosteric BCR::ABL1 inhibitor, STAMP inhibitor targeting the myristoyl binding pocket; y: years.

The overall complication burden was comparable to that of other patients undergoing ALL induction. Patient 2 developed cardiomyopathy between the two episodes, which guided the decision to proceed with CAR T therapy. During the first episode, he experienced methotrexate-induced encephalopathy without seizures, tachycardia secondary to TKI treatment, vincristine-associated neuropathy, and pneumococcal pneumonia. His cardiac function showed a shortening fraction of 30% in 2019, improved to low-normal in 2020, and declined again to 22% at relapse in 2022; treatment with an angiotensin converting enzyme (ACE) inhibitor resulted in full recovery, and his cardiac function is currently normal. Patient 1 relapsed in January 2019 and subsequently developed septic arthritis in January 2020. He also suffered from gastrointestinal bleeding in July 2018, followed by a veno-occlusive disease (VOD)-like complication that was successfully managed with defibrotide, and later experienced several cytomegalovirus (CMV) reactivations during chemotherapy. In summary, patient 2 experienced more complications during the first episode, whereas patient 1 was more affected during the second episode. In both patients, TKI were discontinued and exchanged due to lack of activity in eliminating the BCR::ABL1-positive cells, further supporting the notion that, likely because of their pre-leukemic nature, these cells do not depend on BCR::ABL1 signaling. Blinatumomab was considered but not pursued, as the Ig PCR was negative and therefore no clear indication was present. When asciminib became available, it was initiated as a purely clinical decision, given that BCR::ABL1 levels had not decreased and the clones were not dependent. These experiences highlight the phenomenon referred to as “CML-like” biology in Ph⁺ALL or multilineage involvement, wherein persistent BCR::ABL1 positivity extends beyond the original blast population and involves multiple hematopoietic compartments, with a ‘dormant’ persisting lymphoid BCR::ABL1 clone.¹³ In the two patients reported by us, BCR::ABL1 positivity seemed to be restricted to the lymphoid progenitor compartment and no discordances between BCR::ABL1 transcript levels were noted between bone marrow and peripheral blood. As emphasized in the studies by Zuna and colleagues,¹³ this scenario reflects a stem cell-driven disease in which the BCR::ABL1 clone extends beyond the ALL blast population, and these cells may serve as a reservoir for future disease episodes. Zuna and colleagues have also demonstrated the presence of the BCR::ABL1 clone in T cells, myeloid cells, and normal B cells, reinforcing the concept that some Ph⁺ALL cases may originate in a multipotent progenitor.¹⁷ In both our patients, Ig/TCR-based MRD served as a more definitive predictor of clinical remission status. Unfortunately, additional molecular/genomic testing to compare the additional mutational profile, apart from BCR::ABL1 positivity between the first and second episodes was not feasible as relevant material was not preserved. Mutational profiling of BCR::ABL1 was performed once in each patient: for patient 1, it was negative before initiation of asciminib, and for patient 2, it was negative at the diagnosis of the second leukemia. These children achieved negative Ig/ TCR MRD, remained in morphological remission for several years, and despite persisting BCR::ABL1 positivity showed no immediate disease progression even when TKI therapy was withheld. Of note, although BCR::ABL1 levels remained variably positive, both patients were clinically asymptomatic following their follow-up after the first episode of leukemia. During the first year, they were monitored monthly, and subsequently every 3 months. As they remained without clinical symptoms, diagnostic bone marrow aspirations were discontinued after the first year. At the time of relapse, however, both patients presented with a combination of clinical symptoms, peripheral blood count abnormalities and increasing BCR::ABL1 positivity. As both patients experienced a second BCR::ABL1-positive ALL a significant number of years later, it is evident that even in an apparently indolent state, BCR::ABL1-positive cells leukemias are in a pre-leukemic state which favors clonal evolution and leukemic transformation when additional hits occur. The possibility that further episodes might occur introduces the question whether HSCT is indicated to eradicate the BCR::ABL1-positive reservoir. At present, we have advised the families that a future Ph⁺ leukemia episode may be an indication for HSCT to achieve definitive eradication of the BCR::ABL1 reservoir.

Table 1.Clinical presentation and leukemia-specific findings.

As immunotherapeutic strategies such as CAR T are increasingly integrated into clinical practice, it will become important to distinguish patients with (CD19⁺)-B cell-restricted BCR::ABL1 expression from those with a broader, CML-like involvement - particularly to understand whether BCR::ABL1 can be fully eradicated and if such treatment can be considered definitive, which was not achieved in patient 2.

In conclusion, these observations reinforce the concept that BCR::ABL1-based MRD alone does not consistently predict relapse risk in Ph⁺ ALL. Although some patients may harbor seemingly inactive BCR::ABL1-positive cells for many years without clinical progression but the possibility of later leukemic transformation appears genuine. Identifying clear biomarkers that distinguish quiescent from malignant-prone residual clones would greatly assist in personalizing both the intensity and duration of therapy, including the role of allogeneic HSCT. While immunotherapies and TKI offer clinical benefit, they do not eradicate the underlying pre-leukemic clone. Therefore, prospective follow-up of this subgroup is essential to determine whether outcome differences truly exist and to guide evidence-based, individualized treatment strategies.

Footnotes

• Received July 10, 2025
• Accepted December 17, 2025

Correspondence

*AR-S-S and ES contributed equally as first authors.

^#PMH and CMZ contributed equally as senior authors

CM. ZWAAN - c.m.zwaan@prinsesmaximacentrum.nl

Disclosures

CMZ reports consultancy for Incyte, Kura Oncology, Syndax, BMS, Novartis, and Gilead; institutional trial support from Kura Oncology, Gilead, Jazz, Daiichi Sankyo, Takeda, AbbVie, and Pfizer; advisory or board roles with Novartis, Incyte, and Sanofi; and leadership roles (paid or unpaid) within the ITCC Hem Malignancies Committee, the Dutch MREC Society (Chair), and MREC Utrecht (Chair). ARS serves on a data monitoring committee for a Miltenyi Biotec trial; has consulted for Eusa Pharma and Sanofi; and participated in educational activities for Roche, SERB, and Eusa Pharma. The remaining authors have no conflicts of interest to disclose.

Contributions