IDH1-mutated B-cell acute lymphoblastic leukemia characterized by oncogenic reprogramming of lipid metabolism

B-cell acute lymphoblastic leukemia (B-ALL) is a highly aggressive hematologic malignancy with marked genetic heterogeneity, accounting for approximately 75% of adult ALL cases.¹ The ubiquitous application of high-throughput technologies has facilitated the discovery of novel B-ALL molecular subtypes, including DUX4 and MEF2D rearrangements, which have refined risk stratification for affected individuals.^2,3 Isocitrate dehydrogenase 1 (IDH1) mutations, which occur in approximately 8% of acute myeloid leukemia (AML) patients, are comparatively infrequent in B-ALL.⁴ IDH1 mutations have been well-established as key drivers of leukemogenesis through the production of the oncometabolite (R)-2-hydroxyglutarate (R-2HG).^4,5 Although IDH1-mutated B-ALL had been documented to display distinct transcriptional signatures and prognostic features, this molecular subtype remains understudied due to its rarity.⁶ Previously, we reported a case of IDH1-mutated B-ALL characterized by prominent cytoplasmic lipid droplet (LD) accumulation in leukemic blasts, a morphological phenotype typically associated with ALL-L3 in French-American-British (FAB) classification.⁷ Inspired by this observation, we conducted a comprehensive analysis to characterize features of IDH1-mutated B-ALL in our cohort, which potentially provided novel insights into its biological properties and potential therapies.

In this study, a total of 30 patients with IDH1-mutated B-ALL were enrolled, and an additional 188 patients with IDH1-WT B-ALL were used as controls. All cases with the characteristic immunophenotype of B-ALL were newly-diagnosed and enrolled at the First Affiliated Hospital to Zhejiang University School of Medicine (IIT20240358B). All procedures in studies involving human participants were performed in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments. Written informed consent was obtained from all participants.

IDH1 mutation testing was performed for all patients, using a combination of next-generation sequencing (NGS) and polymerase chain reaction (PCR)-based methods. Among the IDH1-mutated patients, the mutation sites of the IDH1 gene were heterogeneous. Specifically, 13 patients had the R132C mutation, 9 had the R132S mutation, and 3 had the R132G mutation; the mutation sites were unavailable for 5 patients (Figure 1A). Baseline characteristics analysis demonstrated that IDH1-mutated patients were significantly older at diagnosis than with IDH1-wild-type (WT) patients (56 [median, range 26-69] vs. 43 [median, range 14-85] years, respectively; P<0.01). Notably, IDH1-mutated patients showed significantly lower white blood cell (WBC) counts (2.6 [mean, interquartile range, IQR] 2.1-3.0] vs. 9.4 [mean, IQR 3.6-33.0] ×10⁹/L, respectively; P=0.04) and hemoglobin levels than IDH1-WT patients (70 [mean, IQR 57.5-84.5] vs. 89 [mean, IQR 67.0-114.3] g/L, respectively; P<0.01). For genetic alteration, a significantly lower frequency of BCR::ABL1 fusion (0% vs. 34.6%; P<0.01) and a significantly higher frequency of BCOR (13.3% vs. 1.1%; P<0.01) mutation were found in IDH1-mutated patients compared with IDH1-WT patients. Regarding therapeutic outcomes, in comparison with the IDH1-WT group, the IDH1-mutated group exhibited lower complete remission (CR) rates (82% vs. 93%, respectively; P=0.08), higher relapse rates (53% vs. 30%, respectively; P=0.07), and higher mortality rates (32% vs. 16%, respectively; P=0.09) following standard chemotherapy protocols established by Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin⁸ (Table 1). Consistently, IDH1-mutated patients had significantly worse event-free survival (EFS) compared to IDH1-WT patients (median 332 days vs. 446 days, respectively; P=0.02). Meanwhile, patients with IDH1 mutation also tended to have worse overall survival (OS) and relapse-free survival (RFS) (Figure 1B). Through univariable and multivariable analysis, we found that IDH1 mutation was an independent prognostic factor for poor EFS (univariable analysis: HR=1.882 [1.118-3.166]; P=0.017; multivariable analysis: HR=2.126 [1.168-3.867]; P=0.014) (Online Supplementary Table S1). These findings underscored the association between IDH1 mutation and an aggressive clinical phenotype in the B-ALL cohort.

Our initially reported case with IDH1-mutated B-ALL presented with prominent cytoplasmic LD within leukemic blasts at diagnosis; that prompted us to conduct a systematic morphological analysis of B-ALL specimens in our centers. As revealed, cytoplasmic LD were obviously present in IDH1-mutated B-ALL blasts but were absent or minimal (<5%) in IDH1-WT cases in our cohort (Figure 1C). LD are specialized organelles primarily composed of neutral triglyceride (TG) and encased by a phospholipid monolayer,⁹ and the accumulation of LD was reported to be caused by aberrant lipid metabolism, which promoted disease progression and predicted poor outcomes in a variety of cancers.¹⁰ To explore the influence of IDH1 mutation on lipid metabolism in B-ALL, RNA-sequencing (RNA-seq) was conducted to compare the transcriptomic profiles of IDH1-mutated (N=7) with IDH1-WT (N=31) bone marrow blasts. The expression levels of mRNA in sequence data were calculated as RPKM (Reads Per Kilo-base per Million reads).

The criteria of differential gene selection were P<0.05 and |log2(fold change)|>2. Gene Set Enrichment Analysis (GSEA) demonstrated significant enrichment of genes related to oxidative phosphorylation, fatty acid metabolism and adipogenesis in IDH1-mutated cells, indicating a transcriptional reprogramming toward lipid metabolic activation (Figure 1D). Moreover, serum TG and high-density lipoprotein-cholesterol (HDL-C) levels were significantly elevated in IDH1-mutated patients (N=17) compared to IDH1-WT patients (N=177) at diagnosis (Figure 1E). These results indicated an aberrant lipid metabolism in IDH1-mutated B-ALL cells.

Figure 1.Characteristics of IDH1-mutated B-cell acute lymphoblastic leukemia patients. (A) Mutation types of IDH1 gene in our B-cell acute lymphoblastic leukemia patients (B-ALL) patients. (B) Kaplan-Meier analysis of event-free survival (EFS), relapse-free survival (RFS), and overall survival (OS) for IDH1-mutated B-ALL patients. (C) Representative bone marrow morphology of IDH1-mutated (P1-P3) and IDH1-wild-type (WT) (C1-C3) B-ALL patients. Red arrows indicate the locations of prominent lipid droplets. (D) Gene Set Enrichment Analysis (GSEA)for the oxidative phosphorylation, fatty acid metabolism and adipogenesis pathways in IDH1-mutated versus IDH1-WT B-ALL samples. (E) Serum triglyceride (TG) and high-density lipoprotein-cholesterol (HDL-C) levels of IDH1-mutated versus IDH1-WT B-ALL patients at diagnosis. *P<0.05, **P<0.01.

To further explore the correlation between IDH1 mutation and LD as well as TG accumulation, we utilized the LD Assay Kit-Blue from Dojindo Molecular Technologies (Kumamoto, Japan) and the enzymatic TG measurement assay from Applygen Technologies (Beijing, China). Following 24-hour oleic acid (OA, 100 μM) treatment,¹¹ as a positive control, both BALL-1 and RS4-11 cell lines displayed significant increases in LD numbers and TG content (Online Supplementary Figure S1A-C). R-2HG is the well-known oncometabolite produced by IDH1 mutation.^4,5 Strikingly, both R-2HG and IDH1^R132S mutation could significantly augment LD formation and TG biosynthesis (Figure 2A, B and Online Supplementary Figure S1D-G). Moreover, AG120 (ivosidenib, a mutant IDH1 protein inhibitor) treatment significantly reversed the lipid droplet accumulation phenotype (Online Supplementary Figure S1H). R-2HG was reported to inhibit growth of IDH WT AML cells.¹² In contrast to its inhibitory effect in AML cells such as MV4-11, our results demonstrated that R-2HG treatment sustained the proliferation of B-ALL cell lines including BALL-1, REH, and RS4-11 (Online Supplementary Figure S2A-C). Furthermore, we ectopically expressed the IDH1^R132S mutant in B-ALL cells, using pCDH1-MSCV-MCSEF1-GreenPuro plasmid vector for lentivirus-mediated stable transfection, and found that IDH1^R132S overexpression sustained B-ALL cell proliferation and conferred apoptotic resistance under serum-free culture conditions (Figure 2C and Online Supplementary Figure S2D-F). Thus, IDH1 mutation and its oncometabolite R-2HG directly contributed to aberrant lipid metabolism and sustained B-ALL cell proliferation.

To characterize global alterations in lipid-related metabolic profiles, we performed untargeted metabolomic profiling via mass spectrometry in IDH1^R132S-overexpressed and R-2HG-treated RS4-11 cells. Differentially abundant metabolites were identified using a combination of VIP (variable importance in projection) >1 and P<0.05 as being statistically significant (Figure 2D): 60 upregulated and 29 downregulated metabolites were identified in IDH1^R132S-overexpressed cells relative to control cells; 35 upregulated and 108 downregulated metabolites were identified in R-2HG treated cells relative to DMSO-treated ones. KEGG pathway enrichment analysis revealed significant upregulation of glycerophospholipid metabolism in both IDH1^R132S-overexpressed and R-2HG-treated RS4-11 cells (Figure 2E). Glycerophospholipids as well as sphingomyelins are the main components of biological membranes, including LD. As indicated, enhanced glycerophospholipid metabolism was a key metabolic signature of IDH1^R132S-mutated B-ALL, which potentially promoted LD formation.

Collectively, this study provides evidence that IDH1 mutation is a prognostic biomarker predicting poor therapeutic response and adverse clinical outcomes in B-ALL, supporting its role in clinical risk stratification. Previous study by Yasuda et al. identified 7 IDH1/IDH2-mutated patients,⁶ but only 4 of them were included in the subsequent survival analysis. As a result, IDH1/IDH2 mutation as a predictor for poor prognosis was not conclusive enough in their study. By including a larger number of patients, our analysis has provided robust evidence for the role of IDH1 mutation in B-ALL. Therapeutically, small-molecule inhibitors targeting mutant IDH1 (e.g., ivosidenib) have demonstrated efficacy in newly diagnosed IDH1-mutated AML.¹³ In contrast, clinical data on IDH1 inhibitors in B-ALL remain extremely scarce. Our findings position IDH1 mutations and their associated lipid metabolic aberrations as potentially tractable therapeutic targets. It will be critical for future research to evaluate whether integrating IDH1-directed therapy with conventional chemotherapy can improve treatment responses and long-term survival in this patient subset.

Table 1.Baseline characteristics of IDH1-mutated and wild-type B-cell acute lymphoblastic leukemia patients.

Figure 2.IDH1^R132S conferred an oncogenic reprogramming of lipid metabolism in B-cell acute lymphoblastic leukemia patients. (A) Lipid droplets (LD) were stained with Lipi-Blue dye and measured by flow cytometry (FCM). The bar graph represented the mean fluorescence intensity of cells stained with Lipi-Blue dye. (B) Triglyceride (TG) levels were detected in B-cell acute lymphoblastic leukemia (B-ALL) cells with IDH1^R132S mutation. (C) Starvation (remove FBS for 72 hours)-induced apoptosis was detected in IDH1^R132S-overexpressed B-ALL cell lines. The representative FCM analysis of REH cells is shown: gating with the equivalent GFP positivity, cells for the vector control group (N=8,000) and IDH1-mutant group (N=5,000) were analyzed for apoptosis. (D) Dysregulated cell metabolites were analyzed by LC-MS/MS and GC-MS/MS in IDH1^R132S-mutated versus vector cells or R-2HG-treated versus DMSO-treated cells. (E) KEGG pathway analysis was conducted for upregulated cell metabolites in IDH1^R132S-mutated or R-2HG-treated groups were presented. Data are presented as mean ± Standard Error of Mean from three independent experiments (N=3). **P<0.01, ***P<0.001.

A cornerstone of our work is the discovery of pervasive cytoplasmic LD accumulation in IDH1-mutated B-ALL blasts, a previously unrecognized morphological hallmark. We propose a lineage-specific model wherein the oncometabolite R-2HG promotes a net gain in lipid storage by driving lipogenesis that surpasses oxidative consumption. This metabolic reprogramming likely stems from the unique B-cell context, where transcription factors such as PAX5 or constitutive kinase signaling may synergistically amplify lipogenic signals.¹⁴ This stands in contrast to IDH1-mutated AML, where R-2HG often co-operates with mutations that favor oxidative metabolism, accounting for the distinct metabolic phenotypes between these malignancies.¹⁵ Notably, IDH1-mutated B-ALL may converge with ALL-L3 through shared MYC activation, providing a parallel route to LD accumulation.⁷ Functionally, we posit that these LD are not passive reservoirs but active organelles integral to leukemic pathobiology. They may serve as a readily available energy source, modulate membrane fluidity and signaling, and critically, buffer against oxidative stress and ferroptosis, a multifaceted role that represents a compelling avenue for future investigation.

While our findings establish a strong association between IDH1 mutation and LD accumulation, they do not elucidate the underlying causal relationship. Therefore, it remains to be determined whether the aberrant lipid metabolism actively drives LD biogenesis or merely occurs in parallel. Directly testing these competing hypotheses through isotopic tracing and genetic rescue experiments will be an important objective for future work.

Footnotes

• Received June 17, 2025
• Accepted December 22, 2025

Correspondence

*YQ, DS and QL contributed equally as first authors

J. SUN - sunjiehm@zju.edu.cn

X. ZHANG - hillhardaway@zju.edu.cn

J. JIN - jiej0503@zju.edu.cn

Disclosures

No conflicts of interest to disclose.

Contributions

Funding

This work is supported by Key international cooperation projects of the National Natural Science Foundation of China (81820108004, 82200161), the National Key Research and Development Program of China (2022YFC2502700, 2022YFC2502701), the Research Project of Jinan Microecological Biomedicine Shandong Laboratory (JNL-2022034C), and the Natural Science Foundation of Zhejiang Province (MS25H080017).

Acknowledgments

We would like to thank all members of Department of Hematology, The First Affiliated Hospital to Zhejiang University College of Medicine for their support. RNA-sequencing was supported by Oebiotech Company (Shanghai, China) and Acornmed Company (Beijing, China). LC-MS/MS and GC-MS/MS were provided by Shanghai Luming biological technology Co., LTD (Shanghai, China).