Philadelphia chromosome‐positive acute lymphoblastic leukaemia in children and adolescents: A changing treatment landscape and a methodological challenge

Approximately 3%–5% of paediatric acute lymphoblastic leukaemia (ALL) is driven by the BCR::ABL1 fusion gene resulting from t(9;22)(q34;q11.2), known as Philadelphia chromosome-positive (Ph+) ALL. As the treatment landscape of Ph+ ALL has rapidly shifted in the era of chemoimmunotherapy and tyrosine kinase inhibitors (TKIs), this perspective letter brings together experts to review treatment evolution of paediatric Ph+ ALL over the last two decades, highlight ongoing challenges and delineating future directions in optimizing therapy in successor paediatric Ph+ ALL trials. As shown in large ‘Ponte di Legno’ intergroup retrospective studies, before TKIs, the 7-year event-free survival (EFS) and overall survival (OS) of Ph+ ALL children were ~30% and ~40%, respectively, with intensive multi-agent chemotherapy and haematopoietic stem cell transplantation (HSCT) in first complete remission (CR1) offered to all patients with a suitable donor (see references in Supplementary Material).1 An addition of TKIs in subsequent trials led to rapid disease clearance and significant long-term survival improvement. Multiple studies conducted during the past two decades by the European intergroup study on Ph+ ALL (EsPhALL) and/or the Children's Oncology Group (COG) demonstrated significant improvements with 5-year EFS and OS of approximately 60% and 80%, respectively, with continuous TKI-based chemotherapy regimens (Table 1, see references in Supplementary Material).2-6 Importantly, similar EFS and OS rates were observed across trials, regardless of TKI generation (imatinib or dasatinib) and chemotherapy backbone. This precision medicine paradigm also reduced treatment-related toxicity, via omission of prophylactic cranial irradiation for nearly all patients and substantial decrease of HSCT use in CR1. Overall, modern TKI-based therapies have transformed paediatric Ph+ ALL from the once-least favourable subtype of ALL into a more curable entity. Imatinib 340 mg/m2 Post-induction: 18 IT 5 years: 69% (57%–83%) 5 years: 79% (68%–92%) Dasatinib 60 mg/m2 Post day 15 induction: 18 IT 5 years: 60% (46%–74%) 5 years: 86% (76%–96%) Imatinib 300 mg/m2 Post IB: 17 IT + 3 if CNS3 5 years: 57% (49%–65%) 5 years: 72% (64%–79%) Dasatinib 60 mg/m2 Post day 15 IA: 17 IT 5 years: 55% (45%–64%) 5 years: 82% (73%–88%) Modified Total XV-XVI Modified Total XV-XVI Dasatinib 80 mg/m2 Since diagnosis: 19 IT 4 years: 71%a (56%–90%) 4 years: 88%a (81%–96%) Imatinib 300 mg/m2 Since diagnosis: 19 IT 4 years: 49%a (32%–75%) 4 years: 69%a (56%–86%) HSCT in CR1 for children with Ph+ ALL has diminished over the last 20 years from approximately 80% in the EsPhALL2004 trial7 to <10% of patients in the most recent trials (reserved to patients with poor minimal residual disease (MRD) response after 10-week therapy, irrespective of ‘high-risk’ genetics such as IKZF1-plus, whose prognostic relevance remains unclear). The OS of paediatric Ph+ ALL also substantially improved from 40% in the pre-TKI era to 80% with TKI-containing regimens, highlighting post-relapse salvageability in the TKI era. Most patients with relapsed Ph+ ALL can now achieve the second complete remission (CR2) or beyond and proceed to HSCT consolidation (about 30% undergo HSCT in CR2); moreover, the high salvage rates reported in trials do not fully reflect the remarkable recent successes of modern B-cell antigen-targeted immunotherapies, such as blinatumomab, inotuzumab or chimeric antigen receptor T-cell therapies8 (see references in Supplementary Material) which may ultimately improve OS rates further and/or reduce morbidity in post-relapse therapy. Indeed, the incorporation of blinatumomab plus TKI with minimal chemotherapy represents a major paradigm shift in the front-line treatment of adult Ph+ ALL, prompting investigation of analogous strategies in paediatrics. The newly opened international AALL2131/EsPhALL2022 (NCT 06124157, EU CT number 2025-520982-39-00) pilot study with continuous TKI replaces the traditional 8-week consolidation chemotherapy phase with two cycles of blinatumomab and intercalates a third blinatumomab block between subsequent chemotherapy courses (Supplementary Material). This pilot trial aims to establish new baseline outcomes of paediatric Ph+ ALL and to guide future therapeutic options. Use of single-agent TKI ‘maintenance’ therapy in Ph+ ALL patients varies widely across the age spectrum. Most paediatric oncologists stop TKI at 2 years with completion of chemotherapy, although recent identification of the chronic myeloid leukaemia (CML)-like subtype raises the potential for prolonged TKI monotherapy maintenance, similar to approaches used in patients with chronic phase CML. Some paediatric oncologists also use TKI monotherapy after HSCT, for 1 or 2 years as clinically tolerated, although data regarding the potential of this approach in preventing relapse remain limited. In the adult setting, TKI discontinuation has not been standard clinical practice, although very recent data support the feasibility of stopping TKI in patients who did not undergo HSCT (see references in Supplementary Material).9 MRD in Ph+ ALL patients can be monitored using several techniques, including quantitative polymerase chain reaction (qPCR) assessment of BCR::ABL1 fusion (at the transcript or DNA level), qPCR or, more recently, next-generation sequencing (NGS) analysis of immunoglobulin/T-cell receptor (IG/TR) rearrangements or flow cytometry (FC) immunophenotyping. In this context, detectable levels of MRD by any of these techniques may be interpreted by clinicians as signals of disease persistence or recurrence and thereby direct therapeutic interventions. Disease monitoring based on BCR::ABL1 transcript levels is particularly controversial due to its high variability and unclear prognostic significance in paediatric Ph+ ALL. In some cases, BCR::ABL1 qPCR may be frankly positive with concurrent negativity or very low positivity by other methods, consistent with a diagnosis of Ph+ ALL with multilineage involvement (or CML-like ALL) that has been identified in up to one-third of paediatric cases.10 At the current time, it remains challenging to discriminate between primary Ph+ ALL versus CML in lymphoid blast crisis, particularly when the BCR::ABL1 fusion gene encodes for the 210-kD protein (p210). In these cases, physicians may decide to protract TKI treatment because of persistent BCR::ABL1 positivity, similar to the adult Ph+ ALL approach, although growth hindrance remains a relevant concern in paediatric patients. Some experts recommend the monitoring of BCR::ABL1 levels in peripheral blood every 3–6 months post-therapy to detect the early signs of molecular relapse and guide therapeutic decisions.11 However, BCR::ABL1 re-emergence in the absence of overt disease by other MRD assays can be a conundrum for treating physicians, who may opt for a watch-and-wait approach versus initiating TKI monotherapy or multidrug salvage therapy, despite such molecular findings not meeting traditional morphological or newer extended definitions of relapse.12 Similar scenarios may exist when (possibly low) positivity of IG/TR PCR or FC MRD is redetected, suggesting impending relapse (representative case example in Table 2). Notably, children with Ph+ ALL treated with front-line imatinib are more likely to receive dasatinib with second-line salvage chemotherapy upon relapse, even without evidence of emergence of resistance mutations. Newer TKIs (e.g. bosutinib, ponatinib, asciminib, olverembatinib) have also achieved high success in adults with relapsed/refractory Ph+ ALL and are under clinical investigation in children via early phase clinical trials (see references in Supplementary Material).13 Relapse patterns have also changed in the TKI era, with the majority of relapses now occurring after chemotherapy completion compared to ~30% at ≥30 months from diagnosis in the pre-TKI era.2 Indeed, we are aware of three children with Ph+ ALL who relapsed more than 9 years from initial diagnosis (Biondi, Zuna and Loh, personal communication), including one with CML-like disease. Whether prolonged TKI monotherapy would have altered this relapse risk remains unknown. An international ‘Ponte di Legno’ initiative has accordingly been launched to collect real-world data on TKI use after completion of conventional chemotherapy in paediatric Ph+ ALL and to evaluate efficacy of salvage therapies following ‘relapse’ by either standard criteria or by low-level emergence/persistence of MRD. To optimally assess clinical outcomes, details regarding subsequent treatment interventions and prolonged follow-up are needed, which is challenging due to disease rarity and operational burden. Most clinical trials do not routinely capture data after front-line therapy and post-HSCT, partly due to patients being referred to adult institutions for continued care. Robust collection of long-term outcomes may ultimately require a formal registry rather than clinical trials. A more ambitious approach, aimed at optimizing therapeutic plans, calls for adaptive trial designs as in the Sequential, Multiple Assignment, Randomized Trials (SMART),14 which adopt a structured sequence of diagnostic/therapeutic strategies enabling, for instance, the incorporation of innovative treatments for first/impending relapse. The ICH E9(R1) guidelines (see references in Supplementary Material) provide a framework for statistical analysis in the presence of intercurrent events, defined as ‘events occurring after treatment initiation that affect either the interpretation or the existence of the measurements associated with the clinical question of interest’. While the primary estimand (end-point) in Ph+ ALL studies is usually EFS, measured as the time from diagnosis to resistance, relapse, second malignancy or death (whichever occurs first), in future study designs this could incorporate intercurrent events based on additional MRD evaluations during follow-up. This approach could produce a more granular EFS definition, enriching this composite variable with another mode of failure, such as ‘impending relapse’, based on clear time-point(s), cut-offs and MRD assay. For example, the patient described in Table 2 would be considered alive in CR by standard EFS, whereas counted as having an event under this alternative approach. To ensure comparability of results with previous trials, the standard EFS could be added as a sensitivity analysis to the ‘enriched’ EFS. Furthermore, the competing risk approach could be used to estimate the contribution of each EFS component, thus shedding light on the impact of each cause of failure. In conclusion, successful TKI-based therapies during the past two decades have transformed clinical outcomes of children with Ph+ ALL and established a new precedent for precision medicine approaches for other high-risk subtypes of childhood ALL. Recent additional advances in treatment of adult Ph+ ALL (see references in Supplementary Material), particularly intercalation of blinatumomab immunotherapy, reduction of toxic chemotherapy and credentialing of next-generation TKIs,15 challenge paediatric haematologists to investigate similar strategies in children and adolescents. A first step in this direction is the ongoing AALL2131/EsPhALL2022 international study. Efforts should also be made to collect robust information regarding long-term outcomes both from clinical trials and real-world data and to adopt novel statistical approaches to study design and analysis, with the overarching goal of further reducing relapse risk and improving long-term survival with less toxicity. ABa, MGV, ABi, LBS, THT, SKT, SPH, VG and PDL designed the study and wrote the manuscript; all authors reviewed and approved the final manuscript. We thank Valentino Conter for his inspiration and supervision of this paper. We thank Federica Colnaghi for the PCR quantification of BCR::ABL1 transcript of the representative case example. This article was supported by the Italian Ministry of Health within the ‘Ricerca Corrente’ Program. This article was supported by the Italian Ministry of Health within the ‘Ricerca Corrente’ Program. JZ was supported by the Czech Health Research Council NW25-03-00276. JAK and COG operations and Statistics & Data Center are supported by the National Cancer Institute of the National Institutes of Health (U10CA180886, U10CA180899). VG was supported by the Projet Hospitalier de Recherche Clinique-Cancer. GC and the EsPhALL activities in Germany are supported by Deutsche Krebshilfe (#70112958). SKT was supported by the United States National Institutes of Health/National Cancer Institute awards 1U01CA232486 and 1U01CA243072, a Department of Defense Translational Team Science award and a Pennsylvania Department of Health Commonwealth Universal Research Enhancement (CURE) award. SKT is a Scholar of Blood Cancer United (formerly the Leukemia & Lymphoma Society) and holds the Joshua Kahan Endowed Chair in Pediatric Leukemia Research at the Children's Hospital of Philadelphia. SKT receives research funding from Incyte Corporation and Kura Oncology; serves/d on scientific advisory boards for Aleta Biotherapeutics, Amgen, Ascentage Pharmaceuticals, AstraZeneca, C-Further/LifeArc, Kura Oncology, Novartis Pharmaceuticals and Syndax Pharmaceuticals; and has received travel support from Amgen and Jazz Pharmaceuticals (all for unrelated studies). SPH owns common stock in Amgen and has received honoraria from Jazz and Servier. RP has participated in advisory board meetings for Amgen, Novartis, Clinigen and Jazz Pharma and has received honoraria from Amgen for the speaker bureau. GC (or the institution) has received research support or honoraria or travel support from Amgen, Clinigen and Jazz Pharma. The other authors have no conflict of interest. The study was performed according to the Declaration of Helsinki. This manuscript does not report identifiable patient information. The data that support the findings of this study are available from the corresponding author upon reasonable request. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.