Abstract IntroductionB-cell precursor acute lymphoblastic leukemia (B-ALL) in adults is a biologically and clinically heterogeneous disease. Historically, prognostication has relied on so-called standard risk factors, including age, white blood cell count at diagnosis, immunophenotype, and cytogenetic abnormalities. Recently, advances in genomic sequencing have enabled the precise classification of molecular subtypes, including BCR::ABL1-like, DUX4-rearranged (DUX4r), PAX5-altered (PAX5-alt), and ZNF384-rearranged (ZNF384-r) ALL. Additionally, deletion of the IKZF1 gene—encoding a key lymphoid transcription factor—combined with deletions of PAX5 and CDKN2A/CDKN2B, and in the absence of ERG gene deletion, defines the IKZF1plus genotype, which has been linked to poor prognosis, particularly in BCR::ABL1-positive ALL.We evaluated the clinical significance of these recently identified subtypes, in the context of measurable residual disease (MRD) status. Patients and MethodsThe study cohort included 75 adults diagnosed with BCR::ABL1-negative B-ALL. The median age was 44.9 years (range 18–73), and the median leukocyte count at diagnosis was 8.8 × 10⁹/L (range 0.6–635). Immunophenotyping revealed 46 cases (61%) of common B-ALL, 20 (27%) pro-B, and 9 (12%) pre-B. Central nervous system (CNS) infiltration was identified in 9 patients (12%).Patients received an induction treatment either with standard CELL chemotherapy protocols (n=39), or with targeted therapies as part of the Blina-CELL (NCT04554485, n=29) and EWALL-INO (NCT03249870, n=7) clinical trials. Early treatment response was assessed after two blocks of intensive therapy, at week 10 or 11. A poor response was defined as refractory or MRD 1 × 10⁻⁴, an intermediate response as detectable MRD 1 × 10⁻⁴, and a good response as MRD negativity.Standard diagnostic approaches were complemented by transcriptome sequencing, targeted panel sequencing, whole-exome sequencing, and digital MLPA. In addition, extended clinical and statistical analyses were conducted within the CELL. ResultsMolecular subtypes were represented as follows: low hypodiploid 18 (24.0%), BCR::ABL1-like 8 (10.7%), ZNF384-r 7 (9.4%), KMT2A-r 6 (8.0%), hyperdiploid 5 (6.7%), DUX4-r 4 (5.4%), TCF3::PBX1 3 (3.9%), hypodiploid 2 (2.7%), PAX-alt 2 (2.7%), ETV6::RUNX1 1 (1.3%), MEF2D-r 1 (1.3%), MYC-r 1, PAX5 p.P80R 1 (1.3%), ZEB2 (p.H1038R)/IGH::CEBPE1 (1.3%), not otherwise specified 15 (20.0%). IKZF1plus genotype was detected in 14 (20%) subjects (not tested in 4 patients).Five-year overall survival (OS) and progression-free survival (PFS) was 69% (59%–81%) and 60% (50%–73%), respectively. Treatment response was classified as poor in 16 patients (21%), intermediate in 12 (16%), and good in 47 (63%). The highest rates of MRD negativity were observed in patients with the following subtypes: TCF3::PBX1 (100%, n=3), hyperdiploid (80%, n=5), PAX5 p.P80R (100%, n=1), PAX5-alt (100%, n=2), and ZEB2 (p.H1038R)/IGH::CEBPE (100%, n=1).Despite high MRD burden (good response in only 25%, n=4), patients with DUX4-r had excellent overall survival (OS 100%). In contrast, patients with BCR::ABL1-like ALL had both poor MRD response (good response in only 37%) and inferior survival (OS 62.5%). However, there was a trend toward improved OS and PFS in those treated within clinical trials involving frontline immunotherapy.Patients without the IKZF1plus genotype were 4.24-times more likely to achieve MRD negativity at week 10/11 (p=0.027) and showed improved survival outcomes (OS: 76% vs. 58%, p=0.021; PFS: 63% vs. 50%, p=0.10). Patients treated with frontline targeted therapy in the Blina-CELL or EWALL-INO protocols showed a trend toward better molecular response, with a 1.51-times higher probability of achieving MRD negativity compared to those on standard protocols. ConclusionsAdvanced genomic profiling enhances risk stratification by identifying molecular subtypes, including the IKZF1plus genotype, with distinct MRD responses and survival outcomes in BCR::ABL1-negative adult B-ALL. Integration of these findings into clinical practice may improve personalized treatment strategies.
Polívková et al. (2025) studied this question.