Multi-Omics Approaches to Understanding Therapy Resistance in Acute Lymphoblastic Leukemia

Abstract

Therapy resistance remains a major cause of relapse and poor outcomes in acute lymphoblastic leukemia (ALL). Recent multi-omics studies in ALL have revealed that resistance arises from a combination of leukemia-specific genetic lesions, treatment-driven clonal evolution, and adaptive non-genetic programs. Genomic analyses have identified recurrent alterations associated with resistance to chemotherapy, tyrosine kinase inhibitors, and immunotherapies, while single-cell profiling has uncovered heterogeneous cell states that persist during treatment and contribute to minimal residual disease. Emerging epigenetic, proteomic, and metabolic data further indicate that reversible regulatory and signaling changes play a central role in leukemic persistence. Integrative analyses are beginning to define convergent resistance pathways and clinically relevant biomarkers, although longitudinal sampling and clinical translation remain limited. This review summarizes the current multi-omics landscape of therapy resistance in ALL and discusses opportunities to improve risk stratification and therapeutic strategies.

1. Introduction

Acute lymphoblastic leukemia (ALL) is an aggressive hematologic malignancy characterized by the clonal proliferation of immature lymphoid cells in the bone marrow, peripheral blood, and other organs [1]. It represents the most common cancer in children and also occurs in adults, where outcomes are generally less favorable [2]. Over the past several decades, the introduction of intensive multi-agent chemotherapy, targeted therapies such as tyrosine kinase inhibitors (TKIs) for BCR::ABL1-positive disease [3], and novel immunotherapies including blinatumomab [4] and chimeric antigen receptor (CAR) T-cell therapy [5] has markedly improved survival rates, particularly in pediatric populations. However, therapy resistance remains a major obstacle, leading to relapse in a significant subset of patients and contributing to poor long-term outcomes, especially in adults and high-risk subgroups [6].

Therapy resistance in ALL is a multifactorial process influenced by both genetic and non-genetic mechanisms. Genomic alterations such as mutations in NT5C2 [7], CREBBP [8], TP53 [8], and IKZF1 [9] can confer intrinsic or acquired drug resistance. Transcriptional reprogramming, epigenetic plasticity, proteomic changes, and metabolic adaptations further enable leukemic cells to survive under therapeutic pressure [10]. The tumor microenvironment, including bone marrow stromal support and immune evasion mechanisms, also plays a critical role. Given this complexity, single-layer molecular analyses often provide an incomplete picture, limiting the ability to predict resistance or develop effective counterstrategies.

Multi-omics approaches—integrating genomics, transcriptomics, epigenomics, proteomics, and metabolomics—offer a comprehensive framework to unravel the molecular underpinnings of therapy resistance in ALL [10,11]. Advances in high-throughput sequencing, mass spectrometry, and computational biology now allow the systematic profiling of genetic mutations, gene expression programs, chromatin accessibility, protein signaling networks, and metabolic states. Importantly, single-cell multi-omics technologies add spatial and temporal resolution, enabling the dissection of heterogeneous subpopulations and the identification of rare but clinically relevant resistant clones [12]. By integrating data across multiple molecular layers, researchers can uncover convergent pathways and master regulators that might be overlooked by single-modality studies.

In recent years, multi-omics studies have revealed several key insights into ALL therapy resistance. For instance, genomic analyses have traced the clonal evolution from diagnosis to relapse, pinpointing mutations that arise under treatment pressure [13]. Transcriptomic profiling has identified cell states associated with quiescence, stemness, and altered drug metabolism [14], while epigenomic studies have highlighted reversible chromatin modifications that permit adaptive transcriptional responses [15]. Proteomic and phosphoproteomic analyses have mapped rewired signaling cascades, revealing novel kinase dependencies, and metabolomics has shed light on altered energy utilization and redox balance in resistant cells [16]. When integrated, these datasets can identify vulnerabilities that may be exploited therapeutically, such as metabolic bottlenecks or synthetic lethal interactions.

The clinical implications of such integrative multi-omics analyses are substantial. They can enable precise patient stratification, guiding the selection of optimal treatment regimens and informing the use of targeted agents or combination therapies to preempt or overcome resistance [17]. They also facilitate the discovery of predictive biomarkers for minimal residual disease (MRD) monitoring and early detection of resistance emergence, which is critical for timely therapeutic intervention [18]. Furthermore, multi-omics-driven drug discovery can accelerate the identification of novel targets and pathways amenable to pharmacologic modulation.

Despite these advances, several challenges remain in translating multi-omics findings into routine clinical practice. Technical and analytical issues, such as data integration, standardization, and interpretation, must be addressed [19]. Biological heterogeneity within and between patients necessitates large, longitudinal cohorts to capture the full spectrum of resistance mechanisms. Moreover, ethical and logistical considerations, including data sharing, cost, and accessibility of multi-omics technologies, will need to be navigated to ensure equitable implementation [20].

This review aims to synthesize current knowledge on therapy resistance in ALL through the lens of multi-omics research. We first outline the major therapeutic modalities used in ALL and summarize known resistance mechanisms. We then explore how each omics layer—genomics, transcriptomics, epigenomics, proteomics, and metabolomics—has contributed to understanding resistance biology (Figure 1), with particular attention to single-cell and integrative multi-omics approaches. Finally, we discuss the translational potential of multi-omics in guiding personalized therapy, the current challenges hindering clinical adoption, and future directions for leveraging these technologies to improve patient outcomes. By bridging basic and translational research, multi-omics approaches hold promise for overcoming therapy resistance and achieving durable remissions in ALL.

2. Genomic Findings Related to Therapy Resistance in ALL

Genomic alterations are a major driver of both intrinsic and acquired therapy resistance in acute lymphoblastic leukemia (ALL) [21]. Advances in high-throughput sequencing have enabled comprehensive characterization of the mutational landscape, revealing recurrent lesions that promote drug insensitivity, facilitate clonal evolution, and ultimately contribute to relapse [22].

One of the most frequently implicated genomic events in relapsed ALL is mutation of NT5C2 [7], a cytosolic 5′-nucleotidase involved in purine metabolism. Activating NT5C2 mutations enhance the breakdown of nucleoside analogs such as 6-mercaptopurine (6-MP) [23] and thioguanine [24], diminishing their cytotoxicity. These alterations are often absent at diagnosis but emerge under selective pressure from maintenance therapy, highlighting their role in acquired resistance.

Mutations in CREBBP, encoding a histone acetyltransferase, are enriched in relapsed B-ALL and can impair glucocorticoid receptor-mediated transcription, leading to steroid resistance [25]. Similarly, alterations in TP53 disrupt cell cycle checkpoints and DNA damage responses, conferring broad resistance to chemotherapeutics [26]. IKZF1 deletions or mutations, common in high-risk BCR::ABL1-positive and BCR::ABL1-like ALL, are associated with poor treatment response and increased relapse risk [27].

In T-cell ALL (T-ALL), activating mutations in NOTCH1 [28] and loss-of-function lesions in FBXW7 are frequent, and these alterations drive leukemogenesis [29], their contribution to therapy resistance is context-dependent. Relapsed T-ALL often harbors additional lesions affecting cell cycle regulation (CDKN2A/B deletions) and kinase signaling pathways, suggesting stepwise genomic evolution during treatment [30].

Clonal evolution is a hallmark of therapy-resistant ALL. Longitudinal sequencing has shown that resistant subclones may pre-exist at low frequencies before treatment, expanding under therapeutic pressure [31]. In some cases, relapse is driven by a genetically distinct lineage derived from an ancestral pre-leukemic clone, indicating that eradication of dominant diagnostic clones may not prevent disease recurrence [32]. This dynamic evolution underscores the importance of serial genomic monitoring for early detection of resistant clones. Notably, however, several studies [33,34] have demonstrated that the presence of relapse-associated mutations in minor diagnostic subclones does not consistently predict relapse, suggesting that genetic alterations alone are insufficient to determine clonal fitness. Instead, additional factors such as therapy-induced selective pressure, non-genetic adaptive states, and microenvironmental influences play critical roles in enabling subclone persistence and expansion at relapse.

Structural variations also play a role in resistance [35]. For example, CRLF2 rearrangements [36] and JAK-STAT pathway mutations in Ph-like ALL can confer resistance to standard regimens but may create vulnerabilities to JAK inhibitors [37]. Likewise, gene fusions such as ETV6::RUNX1 and KMT2A rearrangements influence treatment response and may shape relapse biology through secondary genomic hits [38].

Recent mutational signature analyses have provided important insights into the evolution of relapsed ALL and the contribution of therapy-induced mutagenesis to resistance. Compared with diagnostic samples, relapsed ALL frequently shows increased treatment-associated mutational signatures, reflecting DNA damage caused by cytotoxic chemotherapy and indicating that therapy itself actively shapes the leukemic genome [39]. Notably, RAS pathway mutations are often enriched at relapse and have been linked to treatment adaptation in pediatric B-cell precursor ALL, supporting their potential role as predictive biomarkers [40].

Of note, the composition of upfront therapy at diagnosis plays a crucial role in shaping the landscape of relapse in ALL. Recent studies have shown that different therapeutic regimens exert distinct selective pressures on leukemic clones, influencing both their survival and clonal evolution. For example, certain therapies may drive the expansion of subclones with specific mutations, while others may induce mutagenic processes that generate novel relapse-associated mutations [40,41]. This highlights the need to consider the therapeutic context when studying relapse, as the initial treatment regimen may significantly impact the genetic and clonal features of subsequent relapses.

Collectively, genomic studies have illuminated diverse mechanisms by which ALL cells evade therapy. While individual lesions can guide targeted interventions, integrative genomic profiling—particularly in serial samples—remains essential for capturing the full spectrum of resistance pathways and informing adaptive treatment strategies.

3. Transcriptomic and Single-Cell Transcriptomic Characterization in ALL

Transcriptomic profiling has provided crucial insights into the molecular programs that underlie therapy resistance in acute lymphoblastic leukemia (ALL) [42]. Bulk RNA sequencing has enabled the identification of gene expression signatures associated with poor prognosis, drug insensitivity, and relapse [43], while single-cell RNA sequencing (scRNA-seq) has further resolved the cellular heterogeneity that often underlies treatment failure [44].

Bulk transcriptomic studies have revealed distinct expression patterns in resistant ALL compared with treatment-sensitive disease. For example, glucocorticoid resistance—one of the most common forms of drug insensitivity in ALL—has been linked to downregulation of glucocorticoid receptor target genes and activation of pro-survival pathways such as PI3K/AKT/mTOR and MAPK signaling [45]. In relapsed B-ALL, upregulation of genes involved in cell cycle progression, DNA repair, and drug efflux (e.g., ABC transporters) has been frequently observed [46]. In T-ALL, transcriptional reprogramming toward a stem-like phenotype, with increased expression of genes regulating quiescence and self-renewal, has been implicated in resistance to chemotherapy and targeted agents [47].

Single-cell RNA-seq has transformed the understanding of therapy resistance by enabling the dissection of heterogeneous leukemic cell populations within individual patients [48]. Resistant subpopulations often pre-exist at diagnosis, characterized by distinct transcriptional states that confer survival advantages under treatment pressure [49]. For instance, rare quiescent or dormant subclones can evade cytotoxic agents targeting rapidly dividing cells, only to expand during remission and cause relapse [50]. In addition, scRNA-seq has revealed transcriptional plasticity, where leukemic cells dynamically shift between cell states in response to therapy, adopting phenotypes that favor persistence and immune evasion [51].

Another critical contribution of single-cell transcriptomics is the ability to map the interplay between leukemic cells and the bone marrow microenvironment [52]. Studies have identified resistance-associated programs driven by stromal interactions, cytokine signaling, and hypoxic niches, highlighting the role of extrinsic factors in sustaining drug-tolerant states.

Integration of transcriptomic data with other omics layers has further clarified the functional impact of genetic and epigenetic alterations on gene expression programs. For example, CREBBP mutations correlate with altered expression of glucocorticoid-responsive genes [53], while IKZF1 deletions disrupt transcriptional networks critical for B-cell differentiation.

Overall, transcriptomic and single-cell transcriptomic analyses provide a high-resolution view of the dynamic and heterogeneous nature of therapy resistance in ALL. These approaches not only illuminate the cellular states and pathways driving relapse but also offer a foundation for identifying biomarkers and therapeutic targets aimed at eradicating resistant subpopulations.

4. Epigenomic Mechanisms Underlying Resistance in ALL

Epigenetic regulation plays a pivotal role in shaping gene expression programs without altering the underlying DNA sequence, and accumulating evidence indicates that epigenomic dysregulation contributes significantly to therapy resistance in acute lymphoblastic leukemia (ALL) [54]. Alterations in DNA methylation, histone modifications, and chromatin accessibility can enable leukemic cells to adapt transcriptionally to therapeutic stress, thereby sustaining survival and promoting relapse.

Aberrant DNA methylation patterns are frequently observed in ALL and can modulate the expression of genes involved in drug metabolism, apoptosis, and cell cycle control [55]. Hypermethylation of tumor suppressor gene promoters, such as CDKN2A/B, silences their expression and contributes to unchecked proliferation [56]. Conversely, hypomethylation at oncogenic loci can sustain the activation of pro-survival pathways. Relapse-specific methylation changes often emerge under treatment pressure, indicating that the epigenome is dynamically remodeled during disease progression.

Histone modifications also play a critical role in resistance. Loss-of-function mutations in the histone acetyltransferase CREBBP, enriched in relapsed B-ALL, reduce histone acetylation at glucocorticoid-responsive promoters, impairing transcriptional activation by the glucocorticoid receptor and conferring steroid resistance [57]. Similarly, mutations in histone methyltransferases such as SETD2 alter chromatin structure and influence DNA repair capacity, enhancing tolerance to chemotherapy-induced genotoxic stress [58].

Chromatin accessibility profiling using ATAC-seq has revealed that resistant leukemic cells frequently maintain open chromatin at enhancers and promoters of survival genes, even under drug treatment [59]. This epigenetic plasticity enables rapid transcriptional reprogramming in response to therapy, allowing cells to adopt drug-tolerant states. Single-cell ATAC-seq studies have further shown that subpopulations with distinct chromatin landscapes pre-exist at diagnosis, potentially serving as reservoirs for relapse [60].

Importantly, epigenomic alterations often act in concert with genetic lesions. For example, IKZF1 deletions disrupt chromatin organization at B-cell regulatory elements, reprogramming transcriptional networks toward a less differentiated and more therapy-resistant state [61]. This interplay underscores the need for integrated genomic and epigenomic analyses to fully elucidate resistance mechanisms.

From a therapeutic standpoint, the reversibility of epigenetic modifications offers a promising avenue for intervention. Agents targeting DNA methyltransferases (e.g., azacitidine) or histone deacetylases have shown potential in re-sensitizing resistant ALL cells to conventional therapies [62]. Future strategies may involve rational combinations of epigenetic modulators with targeted agents or immunotherapies to prevent or overcome resistance.

In summary, epigenomic dysregulation in ALL fosters transcriptional adaptability, enabling leukemic cells to survive therapeutic pressure. Comprehensive mapping of these alterations will be essential for developing strategies to eradicate drug-tolerant populations and improve long-term outcomes.

5. Proteomic and Metabolomic Contributions in ALL

While genomic, transcriptomic, and epigenomic profiling reveal upstream regulatory mechanisms of therapy resistance, proteomic [63] and metabolomic [64] analyses provide direct insight into the functional consequences of these alterations. Proteins and metabolites represent the active effectors of cellular phenotypes, and their abundance, modifications, and interactions often more closely reflect the drug-resistant state than gene-level data alone.

Proteomic profiling, particularly using quantitative mass spectrometry and phosphoproteomics, has uncovered aberrant activation of signaling pathways that promote survival under therapeutic stress [65]. In B-ALL, resistance to tyrosine kinase inhibitors (TKIs) has been associated with upregulation of alternative kinase signaling, such as SRC family kinases [66] or PI3K/AKT/mTOR pathways [67], which bypass BCR-ABL1 inhibition. In T-ALL, proteomic studies have revealed persistent NOTCH1 signaling and elevated phosphorylation of downstream effectors that maintain proliferation despite chemotherapy [68]. Moreover, post-translational modifications (PTMs)—including phosphorylation, acetylation, and ubiquitination—can dynamically alter protein stability, localization, and activity, thereby modulating drug sensitivity [69]. For example, altered phosphorylation states of apoptosis regulators such as BCL2 family proteins can tilt the balance toward cell survival [70].

Metabolomic analyses have highlighted the role of metabolic reprogramming in ALL resistance [71]. Drug-tolerant leukemic cells frequently shift toward oxidative phosphorylation (OXPHOS) to meet energy demands and maintain redox homeostasis. Enhanced mitochondrial metabolism not only supports survival under nutrient or drug stress but also reduces reactive oxygen species (ROS) accumulation, thereby mitigating chemotherapy-induced oxidative damage [72]. Some resistant phenotypes display increased glycolytic flux and glutamine dependence, providing metabolic flexibility that enables adaptation to different microenvironmental conditions. Additionally, alterations in nucleotide biosynthesis pathways can affect the efficacy of antimetabolite drugs, as observed in thiopurine-resistant ALL with increased purine salvage pathway activity.

Integration of proteomic and metabolomic data has revealed coordinated adaptations between signaling and metabolic networks. For instance, activation of PI3K/AKT signaling can drive glycolytic reprogramming [73], while alterations in AMPK and mTOR signaling influence lipid and amino acid metabolism [74]. These interconnected changes create a metabolic landscape optimized for resistance, often at the expense of proliferative speed but with enhanced persistence.

From a translational perspective, proteomic and metabolomic profiling can identify actionable vulnerabilities. Targeting metabolic dependencies with OXPHOS inhibitors or disrupting compensatory kinase signaling pathways has shown promise in preclinical models. Furthermore, proteomic biomarkers, such as phosphorylation patterns or metabolite ratios, may serve as early indicators of resistance emergence, guiding adaptive therapy strategies.

Overall, proteomic and metabolomic approaches complement other omics layers by capturing the functional outputs of resistance mechanisms, offering critical opportunities for biomarker discovery and therapeutic intervention in ALL.

6. Integrative Multi-Omics Analysis and Systems Biology Approaches in ALL

While individual omics platforms provide valuable insights into therapy resistance, no single layer can capture the full complexity of acute lymphoblastic leukemia (ALL) biology. Integrative multi-omics approaches combine data from genomics, transcriptomics, epigenomics, proteomics, and metabolomics to construct a holistic view of resistance mechanisms [75]. This comprehensive perspective is particularly valuable in ALL, where resistance often arises from interconnected genetic, epigenetic, and metabolic adaptations.

Data integration can be achieved through sequential or simultaneous profiling of matched patient samples, enabling direct correlation between molecular alterations at different regulatory levels. For example, combining genomic and transcriptomic analyses can link specific mutations (e.g., CREBBP, NT5C2) to downstream transcriptional reprogramming and pathway activation. Similarly, integration of epigenomic and transcriptomic data has revealed how chromatin remodeling reshapes gene expression in resistant leukemic subclones, while proteogenomic approaches bridge the gap between mutational landscapes and functional protein signaling changes.

Systems biology frameworks further enhance these integrative analyses by modeling molecular interactions within complex networks [76]. Network-based approaches can identify “hub” genes or proteins—master regulators whose perturbation may reverse resistance phenotypes [77]. For instance, multi-omics integration has pinpointed convergent activation of PI3K/AKT/mTOR signaling in resistant ALL, regardless of upstream mutational drivers, suggesting a central vulnerability amenable to targeted inhibition.

Machine learning and artificial intelligence (AI) are increasingly employed to handle the high dimensionality of multi-omics datasets [78]. Predictive models trained on integrated omics features can stratify patients according to relapse risk, forecast treatment responses, identify candidate biomarkers for minimal residual disease (MRD) detection, and facilitate diagnostics and translational discovery [79]. In preclinical studies, AI-driven integration of transcriptomic and metabolomic data has revealed metabolic signatures predictive of glucocorticoid resistance, opening avenues for tailored metabolic interventions [80].

Longitudinal integrative analyses, involving multi-omics profiling at diagnosis, during therapy, and at relapse, provide unique insights into clonal evolution and the temporal dynamics of resistance [81]. Such studies can distinguish early adaptive responses from late, genetically fixed resistance mechanisms, guiding the timing and selection of therapeutic interventions.

Despite these advances, challenges remain in standardizing data processing, harmonizing results across platforms, and validating findings in large, diverse patient cohorts. Nonetheless, integrative multi-omics coupled with systems biology represents a powerful paradigm for unraveling the complexity of ALL resistance. By revealing cross-cutting vulnerabilities and enabling predictive modeling, these approaches hold promise for informing adaptive, personalized therapeutic strategies aimed at preventing or overcoming relapse.

7. Clinical Implications and Translational Applications in ALL

The integration of multi-omics research into the clinical landscape of acute lymphoblastic leukemia (ALL) offers unprecedented opportunities to refine diagnosis, predict therapy response, and personalize treatment strategies [82]. Genomic profiling has already become a cornerstone in risk stratification, enabling the identification of actionable mutations such as BCR-ABL1, JAK-STAT pathway alterations, and IKZF1 deletions that influence therapeutic decisions. Extending beyond genomics, transcriptomic and epigenomic data provide additional layers of patient-specific molecular information, helping to distinguish between primary and acquired resistance mechanisms and to identify minimal residual disease (MRD) signatures with higher sensitivity [83].

Multi-omics insights are driving the development of novel targeted therapies in ALL. For example, the identification of metabolic reprogramming patterns and altered protein interaction networks in resistant clones has informed the design of combination therapies that concurrently target oncogenic signaling, metabolic dependencies, and epigenetic regulators [84]. Such approaches aim to prevent the emergence of drug-tolerant persister cells, a major clinical obstacle. Moreover, multi-omics-derived biomarkers can inform adaptive treatment regimens, allowing early intervention upon detection of resistance signatures through non-invasive liquid biopsies.

The clinical translation of multi-omics research also extends to immunotherapy optimization. Profiling tumor and immune microenvironment interactions at the single-cell level has revealed markers of CAR-T cell exhaustion, immune evasion pathways, and cytokine signaling imbalances, providing a basis for engineering next-generation cellular therapies with improved persistence and efficacy [85]. Epigenomic reprogramming strategies, such as targeting histone modifiers or DNA methyltransferases, are being explored to enhance immune recognition and response durability.

Importantly, the implementation of multi-omics in clinical practice requires robust bioinformatics pipelines and standardized data integration frameworks to ensure reproducibility, scalability, and regulatory compliance [86]. The increasing availability of multi-omics datasets facilitates the development of predictive algorithms and machine learning models capable of delivering real-time, patient-tailored treatment recommendations. However, translation to routine care is challenged by issues of cost, data interpretation complexity, and the need for multidisciplinary expertise.

In summary, multi-omics approaches hold transformative potential for overcoming therapy resistance in ALL by enabling precision oncology strategies that integrate molecular, cellular, and systemic insights. As technological advances reduce barriers to implementation, the future of ALL management will increasingly rely on multi-layered biomarker-guided interventions, bridging the gap between bench discoveries and bedside applications.

8. Future Directions and Challenges in ALL

The application of multi-omics technologies to elucidate therapy resistance in acute lymphoblastic leukemia (ALL) has yielded critical insights, yet several challenges must be addressed to fully translate these findings into clinical benefit (

Table 1. Major multi-omics findings and limitations in therapy resistance of ALL.

Omics	Key Findings	Main Limitations
Genomics	Relapse-associated mutations (e.g., NT5C2, CREBBP, TP53, IKZF1) and structural variants contribute to drug resistance. Relapse often arises from minor pre-existing subclones under treatment pressure.	Genetic lesions alone do not fully predict relapse; limited longitudinal samples; functional relevance of many mutations remains unclear.
Bulk transcriptomics	Resistant ALL shows altered expression of survival pathways, drug metabolism genes, and stemness-related programs.	Bulk data mask cellular heterogeneity; cannot identify rare resistant subpopulations.
Single-cell transcriptomics	Reveals heterogeneous leukemic cell states, including rare drug-tolerant populations that persist through therapy and drive relapse.	High cost; limited patient numbers; challenges for routine clinical use.
Epigenomics	DNA methylation changes, histone modifications, and chromatin remodeling enable transcriptional plasticity and reversible drug tolerance.	Causality difficult to establish; limited single-cell epigenomic data in ALL.
Proteomics	Rewired signaling pathways and altered post-translational modifications directly mediate survival and drug resistance.	Incomplete proteome coverage; scarcity of matched diagnosis–relapse samples.
Metabolomics	Metabolic reprogramming (e.g., increased mitochondrial metabolism) supports survival under therapeutic stress.	Strongly influenced by microenvironment; limited in vivo validation.
Integrated multi-omics	Cross-layer integration identifies convergent resistance pathways and improves relapse risk prediction.	Lack of standardized pipelines; limited external and prospective validation.

). Future research directions should focus on enhancing data resolution, expanding cohort diversity, and integrating multi-dimensional datasets into actionable therapeutic strategies.

First, while current studies have identified numerous genetic, transcriptomic, epigenetic, proteomic, and metabolic alterations associated with resistance, the causal relationships among these layers remain incompletely defined. Longitudinal sampling before, during, and after treatment—coupled with single-cell multi-omics—could reveal the temporal evolution of resistant clones and the interplay between intrinsic tumor features and extrinsic microenvironmental cues. Furthermore, integrating spatial transcriptomics and spatial proteomics may uncover niche-specific resistance mechanisms that bulk profiling fails to detect.

Second, large-scale, harmonized datasets across diverse patient populations are needed to overcome the limitations of small, heterogeneous cohorts. Pediatric and adult ALL differ markedly in molecular landscape and treatment outcomes [49]; therefore, future research must ensure representation across age groups, ethnic backgrounds, and treatment regimens to identify both shared and subgroup-specific resistance mechanisms [87].

Third, the integration of multi-omics data into predictive models remains computationally challenging. Systems biology approaches, incorporating machine learning and network-based modeling, can aid in prioritizing resistance drivers and forecasting therapeutic vulnerabilities. However, these models require rigorous validation in preclinical models and prospective clinical studies to ensure reproducibility and clinical applicability.

Fourth, translating omics findings into therapeutic strategies demands robust functional validation. Genome editing, CRISPR screens, and organoid or patient-derived xenograft models will be essential to confirm the role of candidate targets and assess druggability. Combination therapies designed to simultaneously disrupt multiple resistance pathways hold promise, but optimal regimens must be guided by mechanistic evidence.

Finally, logistical and ethical considerations—including data privacy, cost, and equitable access to omics-guided therapies—must be addressed to ensure that advances benefit all patients. Collaborative consortia, standardized analytical pipelines, and open-access data repositories will be crucial to accelerating discovery and translation.

By addressing these challenges through technological innovation, interdisciplinary collaboration, and clinically anchored research, the next decade holds promise for leveraging multi-omics to overcome therapy resistance and improve outcomes in ALL.

9. Conclusions

Acute lymphoblastic leukemia (ALL) remains a major therapeutic challenge despite remarkable advances in risk-adapted chemotherapy, targeted agents, and immunotherapies. While cure rates for pediatric patients have reached encouraging levels, outcomes for high-risk subsets, relapsed disease, and adults remain suboptimal. Over the past decade, multi-omics technologies—including genomics, transcriptomics, epigenomics, proteomics, and metabolomics—have transformed our understanding of the disease’s molecular underpinnings, revealing the intricate interplay of genetic lesions, dysregulated transcriptional programs, aberrant epigenetic landscapes, altered protein networks, and metabolic reprogramming.

Genomic studies have identified recurrent mutations and structural alterations that not only define biologically distinct subtypes but also illuminate mechanisms of therapy resistance. Transcriptomic and single-cell profiling have uncovered intratumoral heterogeneity and rare subpopulations with stem-like properties that can survive therapy and drive relapse [88]. Epigenomic mapping has revealed reversible regulatory changes that contribute to drug tolerance, offering new avenues for targeted reprogramming. Proteomic and metabolomic approaches have provided functional insights beyond static genetic information, capturing the dynamic states of leukemic cells in response to treatment.

Integrative multi-omics frameworks and systems biology approaches are now enabling the construction of comprehensive disease models that connect alterations across molecular layers, allowing for the identification of convergent vulnerabilities and biomarker signatures with potential clinical utility. These integrated insights are beginning to inform precision medicine strategies, from early risk stratification to the rational design of combination therapies aimed at preventing or overcoming resistance.

Despite these advances, translating multi-omics discoveries into routine clinical practice faces substantial challenges, including the need for standardized methodologies, robust bioinformatic pipelines, and cost-effective assays deployable in diverse healthcare settings [89]. Furthermore, longitudinal sampling and analysis are essential to capture the evolutionary trajectories of ALL under therapeutic pressure, but remain logistically and ethically complex.

In summary, multi-omics research has profoundly reshaped the landscape of ALL biology and therapy, providing an increasingly detailed and actionable blueprint of disease mechanisms. Continued integration of these technologies into translational research, coupled with collaborative efforts across disciplines, holds promise for achieving more durable remissions and, ultimately, cures. The future of ALL management will likely hinge on our ability to leverage these molecular insights to guide individualized treatment, monitor disease in real time, and anticipate resistance before it emerges.