Unleashing the Full Potential of Metabolic Interventions in T-ALL

Summary:

Drugs targeting metabolism have been effectively used in patients with T-cell acute lymphoblastic leukemia (T-ALL) for decades; still, the full therapeutic potential of targeting metabolism has not been completely exploited yet. Here, we highlight the critical need for metabolic biomarkers to advance precision medicine in T-ALL, explore the identification of novel metabolic vulnerabilities, and discuss the potential of targeted therapies and dietary interventions to optimize treatment outcomes.

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive lymphoid neoplasm that predominantly occurs in pediatric and young patients and comprises ∼15% of all pediatric acute lymphoblastic leukemia (ALL) cases. Despite advances in precision oncology and the discovery of leukemia-intrinsic genomic alterations and oncogenic drivers, the treatment of most pediatric T-ALL cases still relies on multiple rounds of intensive combinations of standard chemotherapies (1). Although most pediatric patients with T-ALL achieve remission, these aggressive regimens often lead to significant long-term health complications. In addition, the survival rates among children with high-risk or relapsed T-ALL remain low, at ∼30%. Thus, there is an urgent need to develop novel targeted approaches for T-ALL (1).

Eighty years have passed since Sidney Farber introduced antifolates, marking the first breakthrough in the treatment of childhood ALL. Since then, numerous metabolism-targeting drugs, particularly those targeting nucleotide metabolism, have been approved and incorporated into standard of care for T-ALL, transforming a previously universally fatal disease into one with 90% survival rates. These include purine antimetabolites such as 6-mercaptopurine, pyrimidine antimetabolites such as cytarabine, the asparagine-depleting enzyme L-asparaginase, and antifolates such as methotrexate.

The clinical success of these drugs highlights the therapeutic benefit of targeting metabolism in patients with T-ALL; however, we have not reached the full potential of these metabolism-targeting therapies. Recently, discoveries of metabolic dependencies driving T-ALL progression have reignited interest in metabolic interventions as promising tools for T-ALL management. Here, we discuss the untapped potential of targeting metabolism in T-ALL and propose future directions for subsequent research and therapeutic innovation.

Metabolic Biomarkers: A Path toward Precision Medicine

Prognostic stratification of patients has traditionally relied on their clinical presentation, conventional immunophenotyping, and the characterization of their specific genomic landscape, utilizing standard molecular techniques. In the past 10 years, advances in genomic technologies have provided unprecedented insights into inter- and intrapatient diversity and dramatically improved the accurate classification of ALL, identifying 15 distinct T-ALL subtypes (2). However, metabolic characteristics remain largely underexplored in clinical protocols. Related to this, the epigenetic landscape, mutational status, and transcriptional profiles converge to activate metabolic circuits essential for leukemia cell growth. For instance, NOTCH1-mutated T-ALLs (which encompass more than 60% of cases) are metabolically dependent on the transcriptional programs driven by oncogenic NOTCH1, prominently including metabolic signals from MYC and the PI3K–AKT axis (1). Although specific vulnerabilities across the 15 distinct T-ALL subtypes remain to be decoded, exploring the metabolic dependencies of each subtype is a highly promising avenue. Early T-cell precursor ALL (ETP-ALL), for example, has been shown to rely on the mevalonate pathway for cholesterol biosynthesis, a process essential for leukemia cell growth and survival (3). Thus, it is tempting to speculate that there are potentially yet-unidentified specific metabolic vulnerabilities in other T-ALL subtypes. The integration of multiomic data with the metabolic status of leukemic cells could provide a roadmap for the clinical stratification of vulnerable leukemias to targeted interventions.

Furthermore, current chemotherapies may induce metabolic changes, such as shifts in glycolytic or oxidative phosphorylation (OXPHOS) dependencies, which could be identified by transcriptional and metabolic signatures. When coupled with functional assays to evaluate mitochondrial fitness, these signatures may indicate treatment efficacies or resistance. Biomarkers like these might not only improve diagnostic precision but also inform therapeutic decisions, ensuring tailored interventions while providing a dynamic approach to monitor treatment response over time.

However, identifying ideal and robust metabolic biomarkers requires precise and real-time techniques to track leukemia metabolism. In this setting, techniques such as single-cell metabolomics or CyTOF have the potential to fully delineate actionable biomarkers, providing a more comprehensive understanding of T-ALL metabolism. These advanced methods can generate high-dimensional quantitative data, giving valuable insights into the heterogeneity of leukemic cells and the associated microenvironment.

Another significant opportunity lies in visualizing the fate of metabolites in vivo. We are now seeing a rapid development in strategies for tracking metabolic flux using stable isotopes. Tracer infusion approaches have been optimized to measure rates of glycolysis and the TCA in both preclinical models and in certain human patients with solid tumors (4), yielding invaluable insights as they provide clinically relevant information beyond steady-state measurements. Of note, infusion of ¹³C/¹⁵N-labeled nutrients in patients with T-ALL prior to peripheral blood sample collection would be a minimally invasive approach that might dramatically expand our understanding of the metabolic status and nutrient use of each individual patient’s leukemia, which could lead to better-tailored therapeutic decisions.

Still, current limitations with some of these techniques include low sensitivity, accuracy, and spatial resolution of detected metabolites, as well as artifact inferences, which should all be optimized moving forward. Finally, longitudinal and standardized studies on metabolic biomarkers in patients are essential to uncover how these indicators evolve over time, facilitating their utility to refine risk stratification and monitor therapeutic responses. Real-time metabolic profiling could potentially guide therapy adjustment, predict relapse, and ultimately improve patient outcomes.

Overcoming Therapy Resistance in T-ALL: Metabolic Adaptations and Emerging Strategies

Therapy resistance remains a significant barrier in T-ALL, and it is closely tied to metabolic adaptations. For instance, leukemic cells are known to hijack autophagy in order to survive the toxic insult imposed by certain drug treatments (5). A variety of factors, including different tissue microenvironments and/or leukemia-intrinsic mutational or epigenetic status, might contribute to different metabolic landscapes that promote resistant phenotypes. For example, resistance to treatments in some patients with T-ALL has been linked to mutated genes in metabolic pathways involved in drug metabolism, such as FPGS, PRSP1, and NT5C2 (1). These mutations disrupt critical drug activation or detoxification mechanisms, ultimately compromising treatment efficacy. However, the current understanding of these mechanisms remains incomplete, highlighting the need for more systematic and comprehensive studies. Large-scale genomic analyses in larger cohorts of relapsed/refractory T-ALLs are essential to delineate the full spectrum of mutations in metabolic genes and their functional consequences. Moreover, integrating metabolomic profiling of these samples with genomic data could provide a more holistic view of how these mutations alter cellular metabolism and contribute to treatment failure. Beyond identifying mutations, functional studies are crucial to validate their role in drug resistance and explore new potential metabolic vulnerabilities, paving the way for more effective metabolic mutation-specific therapies.

It is also now well recognized that leukemia metabolism is not uniform even within the same tumor. Importantly, this intratumoral metabolic heterogeneity underscores the need of viewing T-ALL as a systemic disease. Thus, the tissue of origin versus infiltrated organ dichotomy may determine how leukemic cells exploit metabolic routes to sustain leukemia fitness and escape from selective stresses. Indeed, leukemic cells could colonize specialized niches such as the bone marrow, adipose tissue, or meninges to fulfill their metabolic demands while evading chemotherapeutic agents. Stromal cells within these niches may secrete metabolites, including fatty acids and amino acids, which leukemic cells utilize to fuel their energy demands and resist apoptosis. In addition, leukemic cells residing in these environments might adopt a proinflammatory phenotype and exhibit unique metabolic characteristics that promote a chemoprotective metabolic environment. Disrupting the stromal metabolic support by targeting this cross-talk could represent a promising approach to overcome resistance and merits further exploration.

Metabolic flexibility gives leukemic cells the potential ability to switch between glycolysis, glutaminolysis, and OXPHOS. It can also allow them to use alternative carbon sources, depending on environmental cues such as nutrient availability and oxygen levels limited by toxic insults under therapy. This adaptability clearly extends beyond the classical Warburg effect, revealing a complex metabolic landscape. Indeed, the Warburg effect itself has significant implications as the accumulation of lactate helps acidify tumor microenvironments, which can lead to local immune suppression and facilitate cancer progression (6).

However, despite all these insights, the successful progress over the past decade of metabolism-based drugs for leukemia has been limited, with only a few agents advancing beyond early-phase clinical trials and many still in preclinical stages. This challenge is partly attributed to the lack of consideration for metabolic vulnerabilities of noncancer cells within the leukemia microenvironment during the drug discovery process. Moving forward, the development of metabolic therapies must shift from solely assessing cancer cell activity to evaluating their efficacy in immunocompetent or humanized mouse models.

By regulating systemic metabolism, tumor cells can create feedback loops, which drive the production of cytokines, chemokines, and metabolites that promote immune evasion. Understanding how metabolic reprogramming in cancer cells creates immunosuppressive microenvironments can lead to the development of new strategies that enhance the efficacy of immunotherapies. A promising area of research is to identify “metabolic sweet spots,” such as ACLY or glutaminase, which affect cell-intrinsic tumor metabolism (5, 7) and simultaneously control antitumor immunity (8, 9). Reversing the effects of metabolic by-products like lactate or targeting metabolic sweet spots pharmacologically could restore immune function and improve the effectiveness of immune checkpoint inhibitors simultaneously. To this end, the development of clinical-grade inhibitors targeting these routes would be needed.

An additional interesting approach to be further explored would be the use of amino acid–mimetic or metabolite-mimetic compounds that covalently bind to multiple enzymes (such as the glutamine-mimetic DON). Although these compounds lack single-enzyme specificity, their ability to simultaneously target multiple metabolic pathways could be advantageous in limiting the capacity of cancer and stromal cells to rewire their metabolism. This metabolic adaptability, which often confers resistance to single-agent therapies, underscores the potential benefits of combination therapies or agents that simultaneously block multiple pathways.

Disruption of OXPHOS with small molecules or mitochondrial uncouplers has also been extensively evaluated; however, phase I clinical trials for these agents have been discontinued because of severe adverse effects, including lactic acidosis and neurotoxicity (10), calling into question whether this approach can be therapeutically exploited. On the other hand, the repurposing of metformin or statins in combination with chemotherapy regimens in hematologic malignancies highlights the potential to explore thousands of other FDA-approved drugs that have been initially developed for other diseases, such as diabetes, inflammation, or autoimmune diseases, as coadjuvants in the treatment of T-ALL.

L-asparaginase remains part of the first-line treatment for ALL. By depleting extracellular asparagine, this enzyme starves leukemic cells that lack asparagine synthetase activity. Related to this, expanding this approach to include other enzymes that deplete additional circulating amino acids that might be critically needed by leukemic cells could represent a valuable opportunity for future therapies. Recombinant formulations, such as PEGylated forms, could potentially reduce immunogenic reactions and extend the drug half-life, improving the therapeutic index.

Likewise, targeting folate metabolism has been pivotal in the development of modern chemotherapy, and antifolates, such as methotrexate and pemetrexed, have become central components of chemotherapeutic regimens. However, traditional antifolates are associated with significant side effects, emphasizing the urgent need for more selective approaches to disrupt one-carbon metabolism. Focusing on alternative enzymes within these pathways, such as dihydroorotate dehydrogenase (DHODH), serine hydroxymethyltransferases (SHMT1/SHMT2), or phosphoglycerate dehydrogenase (PHGDH), could hold considerable therapeutic potential. Still, lead compounds against these targets require further optimization to enhance their drug metabolism and pharmacokinetic profiles, which are essential for their progression toward clinical application.

Finally, another important challenge to metabolically target T-ALL is that high intracellular levels of certain enzymes might require correspondingly high drug levels to achieve effective target engagement, which leads to additional pharmacokinetic concerns. Targeted protein degradation of those metabolic enzymes using protein degraders could mitigate this problem in the future. Although examples of degrader-based metabolic drugs are currently limited, early efforts targeting transporters may open new avenues for exploring their therapeutic potential. Similarly, advances in drug delivery systems, such as nanoparticle-based formulations, also hold promise for improving the selective targeting of leukemic cell metabolism with reduced drug systemic dosing.

Optimizing T-ALL Treatment through Dietary Interventions

The idea of leveraging diet to complement T-ALL therapy is compelling. Clinical trials are underway to evaluate the feasibility and effectiveness of combining certain dietary interventions with current antileukemic protocols (NCT03157323 and NCT05963971). However, personalized dietary strategies tailored to the metabolic profiles of individual patients are essential to maximize benefits and minimize risks. Malnutrition and obesity, for instance, are associated with adverse outcomes in adolescents and young adults with ALLs. In these patients, an elevated body mass index was associated with hepatotoxicity, hyperglycemia, and worse nonrelapse mortality and overall survival (11). Given these associations, clinical trials such as Improving Diet and Exercise in ALL (IDEAL; NCT02708108 and NCT01317940) have aimed to investigate the effects of caloric or nutrient restriction through diet and exercise (12). However, a central question remains: how to optimize patient metabolic demands while exploiting specific leukemia vulnerabilities?

A promising approach to mimic certain benefits of caloric restriction involves targeted modulation of amino acid intake, with a particular focus on those that are considered essential. It is imperative to distinguish between strategies tested in preclinical models, using cell lines in culture or mouse models, from those intended for translation into clinical practice. Preclinical studies have shown that restricting specific amino acids, such as valine or methionine, can significantly impair leukemia cell growth. For example, T-ALL cells have a strong dependency on valine. In murine models, valine restriction as monotherapy significantly delayed leukemia progression and increased overall survival (13). Interestingly, currently available literature on the effects of limiting other amino acids in leukemia (or in other cancer types) is very limited, underscoring the potential for other yet-unexplored dietary amino acid restrictions to confer significant therapeutic effects.

A different strategy with proved therapeutic potential is fasting. In preclinical models, periodic fasting selectively impairs disease initiation and reverses progression in ALL but not in acute myeloid leukemia (14). This effect is mediated by the upregulation of the leptin receptor, which drives leukemic cells toward terminal lineage differentiation (14). These findings suggest that fasting or its pharmacologic mimetics, such as leptin sensitizers, might benefit patients with advanced ALL. However, its translational impact on human patients remains to be seen. Further research is required to elucidate other molecular mechanisms by which fasting influences the host systemic features. Identifying optimal fasting duration and schedules that maximize therapeutic benefits while minimizing patient discomfort is crucial.

Overall, combining fasting or nutrient-restrictive protocols with targeted therapies, especially immunotherapies, may enhance treatment efficacy. Indeed, distinct dietary components or metabolites produced as by-products of those diets directly affect immunity, providing opportunities to enhance immunotherapeutic strategies. For instance, vitamins and other minerals are essential for the homeostasis of the immune compartment, and deficiencies are associated with reduced numbers of NK cells and macrophages and impaired immune cell functions (15). Addressing these deficiencies through personalized nutrition plans, guided by genomic, transcriptomic, and metabolomic data, may mitigate the immune-suppressive effects of these interventions.

Moreover, dietary interventions have the potential to mitigate treatment-induced toxicities, including gastrointestinal damage, cardiovascular impairments, hepatotoxicity and neurotoxicity, bone diseases, fatigue, and additional hematologic disorders. However, it is still essential to investigate the metabolic consequences underlying these severe side effects. Furthermore, given the emerging potential of microbiome-targeted interventions in modulating metabolism, future studies should explore the interplay between dietary interventions, the microbiome composition of the host, and metabolic therapies in T-ALL.

Despite these promising findings, implementing dietary interventions in clinical practice remains challenging. Factors such as patient compliance, variability in metabolic responses, and potential interactions with existing therapies must be carefully considered. Collaborative efforts among oncologists, nutritionists, and researchers are essential to optimize these strategies for patients with T-ALL. Finally, rigorous clinical trials are needed to evaluate the safety, feasibility, and efficacy of diet manipulations in patients with leukemia, particularly in children, whose dietary interventions must be carefully balanced with growth and development needs. An easier opportunity to implement some of these dietary interventions would be at the time whenever parenteral nutrition of hospitalized patients undergoing treatment is needed. Limiting certain nutrients in this context might be easier to achieve and result in better acute responses to the concomitant chemotherapy.

The future of dietary interventions in leukemia lies in their integration into personalized, multimodal care strategies. By potentially targeting metabolic vulnerabilities, enhancing immune function, and mitigating treatment-related toxicities, these strategies could complement current therapies. Advances in precision medicine, coupled with a deeper understanding of the relationships between diet and leukemia/patient biology, will pave the way for innovative nutritional approaches. Diet interventions would be far more cost-effective and readily available than the introduction of new drugs (with their associated off-target effects and toxicities) in clinical trials. As research continues to unravel the complex interplay between nutrition and leukemia, the potential for dietary strategies to transform pediatric care remains immense. Last but not least, patient education on scientifically grounded, evidence-based information on diet and nutrition is an essential part of public health efforts, especially at a time when scientific information is not fact-checked on all social media platforms.

Authors’ Disclosures

D. Herranz reports grants from NIH/NCI grants R01CA285513 and R01CA236936, The Leukemia & Lymphoma Society Scholar Award 1386-23, The V Foundation Pediatric Translational Grant T2023-024, the Alex’s Lemonade Stand Foundation R Accelerated Award 23-28273, and grants from the Ludwig Cancer Research Princeton Branch. No disclosures were reported by the other author.

Authors’ Contributions

V. da Silva-Diz: Conceptualization, writing, reviewing, editing. D. Herranz: Conceptualization, writing, reviewing, editing.