Detection of IDH1.R132H mutant protein in plasma extracellular vesicles and its application in preoperative noninvasive precision diagnosis of glioma

Abstract

Purpose

Development of a circulating extracellular vesicle-based IDH mutant protein detection technology for diagnosis of IDH mutation status in glioma, enabling personalized resection strategies, precision medication, and disease dynamics monitoring in IDH-mutant glioma.

Methods

We developed a tumor-classification strategy that discriminates tumor types by detecting the IDH1.R132H mutant protein in plasma extracellular vesicles using a bead-assisted, flow-cytometry–based assay. We assessed the differential expression’s capacity to distinguish tumor grades, analyzed pre-operative and post-operative expression variations, and verified consistency with histopathological immunohistochemistry.

Results

The method demonstrated a diagnostic sensitivity of 76.2% (95%CI: 54.9%–89.4%) and a specificity of 86.7% (95%CI: 62.1%–96.3%) for IDH-mutant glioma. Further analysis revealed that the expression level of the IDH mutant protein in plasma EVs was significantly associated with tumor grade in IDH-mutant glioma (P < 0.0001). Moreover, the expression level of the IDH mutant protein in plasma EVs significantly decreased in post-operative patients (P < 0.01) and showed a high degree of consistency with the pathological immunohistochemical results at different tumor grades.

Conclusion

We propose a novel plasma extracellular-vesicle liquid biopsy with potential preoperative application in detecting IDH-mutant glioma, affording noninvasive, rapid, efficient, and low-cost assessment of IDH mutation status.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-026-15810-z.

Introduction

Glioma represent the most common primary malignant tumors of the central nervous system (CNS), accounting for approximately 30% of all brain tumors [1]. The 2021 World Health Organization (WHO) CNS tumor classification established integrated diagnoses that couple histopathologic assessment with molecular markers, thereby cementing the transition of glioma classification to a molecularly informed paradigm [2]. Glioma molecular profiles underpin contemporary diagnostic classification, guide therapeutic decision-making, and refine prognostic assessment. Consequently, molecular characterization is central to precision oncology.

Isocitrate dehydrogenase (IDH) mutation status is central to the integrated diagnostic framework for glioma [3, 4]. It serves as a key molecular biomarker in pathological classification and reporting. IDH mutations primarily occur at codon 132, with the IDH.R132H mutation accounting for approximately 90% of all IDH mutations [5]. According to previous literature, IDH-mutant glioma constitute 40%-50% of all glioma cases, exhibiting a significantly higher 5-year survival rate compared to IDH wild-type glioma [6, 7]. This survival advantage establishes IDH mutation as an independent molecular prognostic marker, operating independently of traditional clinicopathological parameters. Glioma biology is shaped by distinct molecular features. These differences drive variability in growth kinetics and invasiveness between IDH-mutant and IDH-wildtype tumors. Preoperative determination of the glioma molecular subtype is essential for early tumor characterization. This information underpins subsequent precision-treatment strategies. Additionally, the emergence of targeted therapies specifically directed against IDH-mutant glioma is progressively transforming the diagnostic and therapeutic landscape of glioma [8], thereby formally heralding the integration of these tumors into the precision medicine paradigm. Preoperative identification of IDH mutation status is essential. When integrated with the tumor’s spatial relationship to eloquent cortex and critical white-matter pathways, it enables surgeons to plan more evidence-based and individualized treatment strategies.

However, determination of IDH status in routine practice still relies largely on tissue-based molecular pathology assays [9, 10]. These procedures are invasive, not readily repeatable, and time-consuming, limiting their use for longitudinal, noninvasive monitoring.

Therefore, there is an urgent need to develop a non-invasive, rapid, and simple method for IDH mutation detection. This approach would guide personalized resection strategies, precision therapeutics, efficacy evaluation, and dynamic disease monitoring for IDH-mutant glioma. In recent years, liquid biopsy—a non-invasive methodology—has emerged as a promising tool for early glioma diagnosis, particularly through the detection of specific mutant proteins in plasma extracellular vesicles (EVs), showing substantial potential [11–14]. EV-based liquid biopsy assays that detect circulating mutant IDH translate this established biomarker into a noninvasive, serially deployable test. They can support preoperative decision-making and longitudinal surveillance. Broader clinical adoption could advance precision diagnosis and guide treatment selection.

Methods

Application of IDH1.R132H mutant protein detection in blood samples

The overall experimental workflow for the detection of the IDH1.R132H mutant protein from plasma EVs is schematically illustrated in Fig. 1. The workflow integrates peripheral blood EV isolation, synthetic magnetic bead-mediated EV capture, and subsequent flow cytometry analysis (Fig. 1). Technical validation utilized 90 plasma samples stratified into three cohorts: 30 IDH1.R132H-mutant glioma, 30 IDH1 wild-type glioma, and 30 healthy controls. Subsequently, 35 glioma patients were double-blinded and randomly enrolled as a validation set for further verification.

Fig. 1.

Construction and research framework for the IDH1 mutant protein detection system

Cell lines

The following human cell lines were used: The U87 MG cell line stably expressing the IDH1.R132H mutant was obtained from the Cell Bank of the Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China). Human Astrocytes (HA-1800) were provided by the Chinese Academy of Medical Sciences and Peking Union Medical College. For IDH cells, the basal medium was Minimum Essential Medium (MEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco), 1% (v/v) penicillin-streptomycin solution, 1.5 g/L sodium bicarbonate, and 0.11 g/L sodium pyruvate. HA cells were cultured in astrocyte-specific basal medium (ScienCell) supplemented with 2% (v/v) FBS and 1% (v/v) astrocyte growth supplement.All cells were maintained in a humidified incubator (Corning, 430167 culture dish) at 37 ℃ with 5% CO₂. The culture medium was refreshed every 48 h. Subculturing was performed when cells reached 80–90% confluence. IDH cells were dissociated using 0.25% (w/v) trypsin, while HA cells were treated with 0.05% (w/v) trypsin for passaging.

Study cohort

This study received approval from the First Medical Center of the Chinese PLA General Hospital (Approval No. S2018-268-02).The IDH-mutant glioma cohort included cases of both major molecular subtypes according to the WHO 2021 classification: astrocytoma, IDH-mutant and oligodendroglioma, IDH-mutant and 1p/19q-codeleted. Following the Institutional Review Board-approved protocol, 95 patients with glioma and 30 healthy donors—all free of serious illnesses and without surgical history in the past 12 years—were enrolled. Informed consent was obtained from all participants. Preoperative blood samples were collected on the day of surgery, while postoperative blood samples were collected within one week post-surgery. Blood was drawn into vacuum tubes and centrifuged at 2500 g for 10 min to isolate plasma. Plasma specimens were then stored at -80 °C.The IDH status was determined by Immunohistochemistry and genotyping on tumor DNA with amplification by polymerase chain reaction (PCR) of the exon 4 (including codon 132) of the IDH1 gene followed by direct sequencing.

EVs isolation and extraction

Ultracentrifugation was employed for EV extraction. Each cell line utilized 15 culture dishes to obtain sufficient EVs. When cell confluence reached 75–80%, the medium was replaced with exosome-depleted fetal bovine serum (FBS-EV-free) supplemented EV-specific production medium, followed by 24 h of continued culture before supernatant collection; at this stage, each dish contained approximately 3 × 10⁶ cells. From 15 dishes, approximately 80 mL of culture supernatant was collected and subjected to sequential centrifugation: 800×g for 5 min (to remove large dead cells) and 2000×g for 10 min (to eliminate large cellular debris). The resulting supernatant was filtered through a 0.22 μm pore filter (Millipore, USA) to exclude large vesicles. The final supernatant underwent ultracentrifugation (Beckman Optima XPN-100 centrifuge, 70 Ti rotor, Beckman-355618 tube type, k-factor = 44, no brake) at 100,000×g for 2 h to pellet small vesicles. The pellet was resuspended in 26 mL of pre-chilled PBS and centrifuged again at 100,000×g for 2 h to remove contaminating proteins. After discarding the supernatant, EV pellets were resuspended in PBS, aliquoted, and stored at 4℃(short-term) or -80 °C (long-term). All centrifugation steps were performed at 4 °C.

To isolate EVs from patient plasma, 200µL of plasma was thawed on ice. The sample underwent sequential centrifugation at 4 °C: first at 800×g for 5 min to remove cells, followed by 2000×g for 15 min to eliminate large cellular debris. The resulting supernatant was filtered through a 0.22 μm pore filter to exclude vesicles larger than 200 nm. Plasma was then subjected to overnight ultracentrifugation at 150,000×g for at least 8 h at 4 °C. The supernatant was discarded, and the pellet adhering to the centrifuge tube wall was resuspended in 26 mL of pre-chilled PBS. A subsequent centrifugation at 100,000×g for 2 h at 4 °C was performed to remove protein contaminants from the EV surface. Finally, EVs were resuspended in PBS, aliquoted, and stored at 4 °C (short-term) or -80 °C (long-term).

Transmission electron microscopy

A 10 µL aliquot of extracellular vesicle suspension was applied onto a 200-mesh copper grid support film (Beijing Zhongjing Keyi Technology Co., Ltd., China) and left to adsorb for 2 min. Subsequently, 10 µL of 1% (w/v) uranyl acetate solution was added for negative staining. After 30 s, excess stain was aspirated. Morphological characterization and analysis were performed using a Hitachi HT7700 transmission electron microscope (Hitachi High-Technologies Corporation, Japan).

Nanoparticle tracking analysis

Concentration and size distribution of extracellular vesicles were quantified using a Malvern NanoSight LM14 nanoparticle tracking analyzer (Malvern Panalytical Ltd., UK). During experiments, samples were injected via a syringe pump system. The integrated scientific-grade CMOS camera (gain setting: 16; detection threshold: 7) performed dynamic imaging analysis, with each acquisition lasting 60 s. To prevent cross-contamination, the measurement chamber was thoroughly rinsed three times with phosphate-buffered saline (PBS) between samples. Raw data were collected and processed using the accompanying NTA software (version 2.3; Malvern Panalytical Ltd., UK), yielding particle size distribution curves and quantitative concentration data.

Flow cytometry analysis

EVs isolated from cell culture supernatants or plasma were conjugated with aldehyde-modified latex microspheres (Thermo Fisher Scientific; diameter: 4 μm; catalog no. A37304; concentration: 1.0 × 10⁷ beads/mL) at a ratio of 4 µg EVs per 1 µL microspheres (equivalent to 1.0 × 10⁴ microspheres). The coupling was performed at room temperature using a thermoshaker (Thermo-Shaker MS-100; 220 rpm/min) with continuous rotation for 10 min. Subsequently, 100 µL of phosphate-buffered saline (PBS) was added to the reaction system, followed by overnight incubation (≥ 7 h) at 4 °C with vertical rotation. The next day, 5% bovine serum albumin (BSA) was added, and the mixture was rotated continuously at room temperature using the same thermoshaker (220 rpm/min) for 1 h to terminate the conjugation reaction. The microspheres were then resuspended and washed with PBS containing 2% BSA, centrifuged at 2000×g for 1 min at 4 °C, and the washing step was repeated twice.

A 100 µL aliquot of primary antibody working solution (anti-IDH antibody, Abcam; catalog no. ab230949; dilution 1:50) was added to the microsphere system. The mixture was incubated at room temperature with continuous rotation (220 rpm) for 1 h. Following primary antibody incubation, the washing procedure described above was repeated twice.Subsequently, 100 µL of fluorescently labeled secondary antibody working solution (Alexa Fluor 488-conjugated goat anti-rabbit IgG, Immunoway; catalog no. RS3211; dilution 1:2000) was introduced. The secondary antibody incubation was performed at room temperature in the dark with continuous rotation (220 rpm) for 30 min, followed by two additional washing steps.A negative control group subjected only to secondary antibody incubation was included to validate specificity. Signal detection and data analysis were conducted using a BD Accuri C6 flow cytometer (BD Biosciences), with instrument parameters optimized for Alexa Fluor 488 excitation/emission wavelengths.The gating strategy was established as follows: First, the main population of EV–bead complexes was identified and gated based on forward scatter (FSC) and side scatter (SSC) characteristics to exclude debris. Next, doublets and aggregates were excluded using an SSC-Height versus SSC-Area dot plot to isolate singlet beads. Finally, the mean fluorescence intensity (MFI) of the singlet population was analyzed in the Alexa Fluor 488 channel (Supplementary Figure S1).

Statistical analysis

Flow cytometry data were analyzed using GraphPad Prism version 10.1.2 (GraphPad Software, Inc.). Multivariate binary logistic regression analysis was performed to evaluate the independent predictive value of MFI. Receiver operating characteristic (ROC) curve analysis was employed to evaluate the specificity and sensitivity of IDH-mut extracellular vesicles (IDH + EVs) and their mean fluorescence intensity (MFI) in distinguishing IDH-mut glioma from healthy controls and IDH-wild glioma. The area under the ROC curve (AUC) was calculated for each biomarker. All ROC analyses were performed with GraphPad Prism 10.1.2 (GraphPad Software, Inc.). The cutoff value was determined using the Youden index. Statistical significance was defined as p < 0.05.

Results

Patient characteristics

Table 1 summarizes the baseline demographic and clinical characteristics of enrolled cohorts. The IDH-mutant glioma cohort included cases of both major molecular subtypes according to the WHO 2021 classification: astrocytoma, IDH-mutant and oligodendroglioma, IDH-mutant and 1p/19q-codeleted. The study cohort comprised a training set and an independent validation set. The training cohort initially comprised 90 individuals, including 30 (33.3%) IDH-mutant cases. The subsequent blinded validation cohort included 35 patients, with 21 (60.0%) confirmed as IDH-mutant. Baseline characteristics, including age, gender, and disease course, showed no statistically significant differences between the training and validation cohorts (P > 0.05). Within the training cohort, IDH-mutant cases (n = 30) exhibited a higher proportion of WHO grade 3 tumors (16 cases, 53.3%) compared to grade 2 (14 cases, 46.7%). Similarly, the validation cohort’s IDH-mutant cases (n = 21) also showed a predominance of WHO grade 3 tumors (11 cases, 52.4%), with fewer grade 2 cases (10 cases, 47.6%). Notably, in the training cohort, the IDH-mutant subgroup showed significantly higher Karnofsky Performance Status (KPS) scores compared to the IDH-wildtype subgroup (P < 0.05). All patients in both cohorts underwent craniotomy following admission, consistent with standard clinical management protocols.

Table 1.

Baseline demographics and clinical characteristics of enrolled cohorts

Characteristic	Number	Percentage (%)
Age, years	48.47 ± 1.23
Gender
Female	60	48.0
Male	65	52.0
Integrated Pathology
Astrocytoma, IDH-mutant	34	35.8
Oligodendroglioma, IDH-mutant and 1p/19q-codeleted	17	17.9
Glioblastoma, IDH-wildtype	44	46.3
WHO grade
2	24	25.3
3	27	28.4
4	44	46.3
In all cohorts
IDH-mutant	51	40.8
IDH-wild	44	35.2
Healthy controls	30	24.0
Disease course, months	8.12 ± 0.80
KPS score	77.89 ± 1.05

High expression of mutant protein in extracellular vesicles derived from IDH1.R132H glioma cells

To assess the presence and detectability of the IDH1.R132H mutant protein within cell-derived EVs, we developed a magnetic bead-based EV analytical platform. EVs were isolated from conditioned media of HA and IDH1.R132H-mutant glioma cells via ultracentrifugation. Transmission electron microscopy (TEM) confirmed typical cup-shaped membrane structures (Fig. 2A), while nanoparticle tracking analysis (NTA) demonstrated a particle size distribution concentrated at 144.6 ± 49.9 nm with a concentration of 7.89 × 10⁹ particles/mL—physicochemical properties consistent with exosomes (Fig. 2B).Isolated EVs were coupled to aldehyde-latex microspheres, followed by immunolabeling with an IDH1.R132H mutation-specific monoclonal antibody. Subsequent incubation with fluorescent secondary antibodies and flow cytometry enabled quantitative detection of mutant proteins on the EV surface. Results revealed significantly higher fluorescence signal intensity in EVs derived from IDH1-mutant glioma cells compared to normal HA-derived EVs (P < 0.001; Fig. 2C-D). This differential was consistently observed in patient plasma samples, confirming that circulating EVs stably carry tumor-specific mutant protein biomarkers.

Fig. 2.

Magnetic Bead-Based Isolation and Detection of IDH1 Mutant Protein in EVs from Glioma Cell Lines. A Transmission electron microscopy (TEM) reveals typical cup-shaped morphology of EVs derived from glioma cells (scale bar = 200 nm). B Nanoparticle tracking analysis (NTA) demonstrates particle size distribution characteristics of EVs in glioma cell culture supernatant. C Representative flow cytometry histograms of IDH1 mutant protein-positive EV-microbeads from different cell lines (top: HA cell line; bottom: IDH1-mutant cell line). D Quantitative comparison of mean fluorescence intensity (MFI) of IDH1 mutant protein in EVs isolated from normal human astrocytes versus IDH1-mutant glioma cells

IDH1.R132H mutant protein in plasma EVs as a diagnostic biomarker for glioma

To systematically evaluate the clinical utility of plasma EVs in detecting IDH1 mutations, we established a rigorous validation framework. Nanoparticle tracking analysis (NTA) quantified EV concentrations across clinical cohorts: healthy controls (n = 9.5 ± 8.4 × 10⁷ particles/mL), IDH1-mutant glioma (n = 4.4 ± 2.2 × 10⁸ particles/mL), and IDH1 wild-type glioma (n = 7.6 ± 5.4 × 10⁸ particles/mL). The mutant group exhibited significantly elevated EV concentrations versus healthy controls (P < 0.05), while no significant difference was observed between mutant and wild-type groups (P = 0.25; Fig. 3A).Capture efficiency optimization employed volume saturation experiments. Healthy plasma EV preparations (0–250 µL) were co-incubated with 5 µL aldehyde-functionalized magnetic beads, followed by flow cytometric quantification using DIO dye for EV membrane labeling. EV adsorption plateaued when EV volume exceeded 200 µL. Given glioma patients’ elevated plasma EV concentrations, a fixed 200 µL plasma-to-5 µL beads ratio was standardized to maximize capture efficiency in clinical samples (Fig. 3B).

Fig. 3.

Performance Evaluation of the IDH1 Mutant Protein Detection System in Clinical Plasma Samples. A Detection of EV concentration across different groups. B Volume saturation curve of magnetic bead-captured plasma EVs. C Validation of antibody efficacy for IDH1 mutant protein detection in plasma. D Flow cytometric detection of IDH mutant protein in EVs from different clinical plasma sample groups. E ROC curve analysis of clinical plasma samples. F Diagnostic test results from a blinded cohort evaluation

To validate the targeted binding capacity of the IDH1.R132H mutation-specific monoclonal antibody in complex plasma matrices, Recombinant IDH1.R132H mutant protein (5 µg) and IDH1 wild-type protein (5 µg) was separately added to healthy human plasma, followed by magnetic bead-based capture of EVs and flow cytometric detection. Results demonstrated a 6-fold increase in mean fluorescence intensity (MFI) in the mutant protein-spiked group compared to the wild-type group (95% CI: 5020–13249, P < 0.01). Crucially, no cross-reactivity was observed with the protein-free plasma control (Fig. 3C), confirming the antibody’s high specificity in complex biological matrices.

To systematically evaluate the clinical translation potential of the plasma-based IDH1 mutant protein detection system, this study established a clinical sample cohort comprising 30 IDH1-mutant glioma, 30 IDH1 wild-type glioma, and 30 healthy controls. Rigorous controlled experiments validated the diagnostic efficacy of the plasma-based IDH mutation detection technology. To exclude the potential influence of confounding factors, we performed a multivariate logistic regression analysis adjusting for age and tumor volume. The results demonstrated that plasma EV MFI remained an independent predictor of IDH mutation status (OR = 1.098; 95% CI: 1.060–1.138; P < 0.001), while age and tumor volume showed no statistically significant association (Supplementary Table S1). Statistical analysis revealed significant differences between the IDH1-mutant group and control groups (P < 0.0001), confirming its utility as a non-invasive adjunct for molecular subtyping of glioma (Fig. 3D). Receiver operating characteristic (ROC) curve analysis demonstrated an area under the curve (AUC) of 0.933 (95% CI: 0.87–0.99, P < 0.0001) for detecting IDH1 mutations.Using histopathologically confirmed cohorts (30 IDH1-mutant and 30 wild-type cases), the optimal diagnostic threshold was determined via maximum Youden index calculation (Youden index = 0.77) as 3541, achieving 93.3% sensitivity (95% CI: 78.7%–98.8%) and 83.3% specificity (95% CI: 66.4%–92.7%) (Fig. 3E). It should be noted that this cutoff value is assay-specific and dependent on the particular instrument settings used in this study. To control for inter-batch variability, all samples were processed using a strictly standardized protocol, and data were acquired using identical flow cytometer parameters (gain and voltage) to ensure consistency across different runs. Further validation through a blinded cohort (n = 35) revealed high concordance with histopathological gold standards (χ² = 8.241, P = 0.004). At this threshold, the diagnostic sensitivity was 76.2% (95% CI: 54.9%–89.4%), and the specificity was 86.7% (95% CI: 62.1%–96.3%) (Fig. 3F).

EV liquid biopsy enables noninvasive molecular grading and postoperative monitoring in IDH1-mutant glioma

We next investigated whether the level of IDH1 mutant protein in plasma EVs could differentiate between histological grades. To this end, we pooled our IDH-mutant cases into a group representing lower histological grade (CNS WHO grade 2, combining grade 2 astrocytomas and oligodendrogliomas, n = 20) and a group representing higher histological grade (CNS WHO grade 3, combining grade 3 astrocytomas and oligodendrogliomas, n = 29).This magnetic bead-based EV platform effectively differentiated WHO grade 2 from WHO grade 3 IDH1-mutant glioma (P < 0.0001), offering a potential non-invasive approach to assist in tumor grade stratification. A positive correlation between Ki-67 proliferation index and tumor aggressiveness was confirmed, with significantly higher Ki-67 expression in WHO grade 3 IDH1-mutant glioma versus grade 2 (P < 0.05).To dynamically monitor postoperative IDH1 mutant protein fluctuations, we systematically collected plasma samples 7 days post-surgery from 8 patients with complete follow-up. Results demonstrated a significant reduction in IDH1 mutant protein levels in postoperative EVs compared to preoperative baseline (P < 0.01), suggesting potential for longitudinal treatment monitoring. Furthermore, two IDH1-mutant glioma patients covering WHO grades 2–3 and one IDH wild-type glioma patient were randomly selected for analysis. Results demonstrated that mean fluorescence intensity of IDH1 mutant protein in plasma EVs from WHO grades 2–3 patients showed high concordance with matched tumor tissue IHC staining results (Fig. 4D).

Fig. 4.

Multivariate Analysis of Plasma Extracellular Vesicles (EVs) in IDH1 Mutant glioma. A Differences in IDH1 mutant protein levels in plasma EVs across WHO-graded glioma patients. B Variations in Ki-67 expression levels in plasma EVs among patients with different WHO grades. C Longitudinal comparison of preoperative and postoperative IDH1 mutant protein expression in plasma EVs. D Correlation between protein expression in EVs from different-grade glioma and immunohistochemistry (IHC) staining (Panel 1: Protein levels in EVs from plasma samples of IDH1 wild-type glioma; Panel 2: Protein levels in EVs from plasma samples of WHO grade 2 IDH1-mutant glioma; Panel 3: Protein levels in EVs from plasma samples of WHO grade 3 IDH1-mutant glioma)

Discussion

Glioma represents the most prevalent and highly malignant primary brain tumor, with its incidence varying among different populations [15]. The IDH1 mutation is widely recognized as a crucial biomarker for glioma, particularly in grade II glioma, where its presence is closely linked to patient prognosis [16]. Existing clinical assays for IDH1 mutations remain invasive, difficult to repeat, and suboptimally sensitive [17, 18]. We describe a magnetic bead-assisted plasma EV assay that is noninvasive, rapid, amenable to serial testing, and cost-efficient. The platform integrates with standard flow cytometry, enables preoperative molecular grading and postoperative surveillance, and shows concordance with tissue immunohistochemistry. Together, these features support earlier diagnosis and more individualized management of glioma. By coupling EVs to micron-scale microspheres, the method enables reliable flow-cytometric analysis of these nanoscale particles, overcoming the size and light-scatter limitations of conventional cytometers. The system achieves signal amplification and single-particle resolution through covalent conjugation between aldehyde latex microspheres and EV membrane proteins.

While prior research primarily utilized cell-free DNA (cfDNA) as a candidate biomarker for assessing IDH1.R132H mutation status in liquid biopsy, its sensitivity remained constrained (25%–75%) [19–23]. Through technological optimization, our cohort analysis achieved 66.7% sensitivity (95% CI: 38.7%–87.0%), 92.3% specificity (95% CI: 62.1%–99.6%), and an AUC of 0.886 (P < 0.001). In conclusion, plasma EV–based monitoring offers noninvasive, serial evaluation of tumor-secretome changes and is especially useful for deep-seated or multifocal disease. Despite suboptimal sensitivity, its speed, repeatability, and workflow compatibility provide a practical foundation for dynamic risk modeling in glioma management.

In the analysis of correlations with multiple disease parameters, IDH1 mutant protein levels were significantly higher in WHO grade 3 glioma compared to grade 2, a grade-specific expression pattern potentially associated with blood-brain barrier (BBB) disruption [24]. The study observed that EV concentrations in peripheral blood were higher in tumor patients than in healthy individuals, likely due to tumor-induced BBB disruption leading to increased EV leakage [25]. High-grade glioma exhibit more severe BBB disruption, explaining the elevated IDH1 mutant protein levels compared to lower-grade tumors.Furthermore, the Ki-67 proliferation index correspondingly increased in high-grade IDH1-mutant glioma [26], suggesting a correlation between IDH1 mutant protein expression and tumor proliferative activity, thereby reinforcing our findings. This provides critical evidence for precision oncology in glioma management, particularly in high Ki-67 index subgroups where elevated IDH1 mutant protein may reflect its prognostic significance in tumor progression. For postoperative monitoring, IDH1-mutant protein levels in plasma EVs declined significantly by postoperative day 7, whereas radiographic signs of treatment response typically emerged only 1–2 weeks after surgery [27]. Furthermore, we performed a correlation analysis between the difference in tumor volume before and after surgery and the expression level of the IDH mutant protein. The results indicated no statistically significant association between the two variables (r = 0.015, P > 0.05). This finding suggests that surgery-related factors may influence the dynamic expression of the target protein postoperatively, the specific mechanisms of which warrant further investigation. Existing research confirms that tumor-derived extracellular vesicles play pivotal roles in intercellular signaling, influencing tumor growth and metastasis [28]. Monitoring plasma IDH1 mutant protein fluctuations thus enables dynamic assessment of tumor burden and informs personalized treatment strategies. For instance, postoperative reductions in IDH1 mutant protein levels may indicate decreased tumor load, thereby guiding subsequent therapeutic decisions and prognostic evaluations.Concurrent findings revealed that IDH1 mutant protein expression in plasma EVs across different-grade glioma consistently matched IHC results from paired tumor tissues, further confirming that plasma EV detection can non-invasively reflect molecular pathological characteristics of IDH1 mutations in tumor tissues.

A limitation of this study is that we were unable to directly compare the level of the IDH1.R132H mutant protein in plasma EVs with that in EVs isolated from matched tumor tissues, primarily due to the practical challenges in obtaining sufficient fresh-frozen tumor tissue for EV isolation from the same surgical cohort. Additionally, regarding the technical validation, while we optimized the bead-capture system using saturation curves to ensure signal stability, the absolute coupling efficiency was not explicitly quantified in this study. This represents a methodological limitation that warrants further investigation in future technical refinements. Moreover, the limitations of this study primarily lie in the constraints of sample size and the inadequacy of clinical validation. Despite the modest cohort size, magnetic bead–assisted flow cytometry enabled reliable detection of the IDH1 R132H mutant protein. However, the absence of large, multicenter validation may limit generalizability, highlighting the need for broader prospective studies. Furthermore, because > 80% of cases have not yet reached the mortality endpoint, survival estimates are not yet reliable. Comprehensive analyses of overall survival will be conducted after adequate events accumulate. Therefore, future research should involve expanding the sample size and conducting long-term longitudinal studies to better validate the discoveries of this study and explore the potential influences of various factors on the expression of the IDH1 mutant protein, thereby enhancing the feasibility of its clinical application.

Conclusion

In this study, we developed a non-invasive preoperative detection workflow for the glioma IDH1 R132H mutation based on peripheral circulating EVs. Our findings suggest that this method offers a potential approach for the rapid, efficient, and cost-effective qualitative and quantitative analysis of IDH1 mutation status. This technology holds promise for supporting future clinical workflows—including personalized surgical strategies, precision medication regimens, therapeutic efficacy assessment, and disease progression monitoring—thereby contributing to the advancement of precision oncology.

Abbreviations

EVs

Extracellular vesicles

IHC

Immunohistochemistry

CNS

Central nervous system

WHO

World health organization

IDH

Isocitrate dehydrogenase

HA

Human astrocytes

FBS

Fetal bovine serum

PCR

Polymerase chain reaction

BSA

Bovine serum albumin

ROC

Receiver operating characteristic

AUC

Area under curve

KPS

Karnofsky performance status

TEM

Transmission electron microscopy

NTA

Nanoparticle tracking analysis

MFI

Mean fluorescence intensity

FSC

Forward scatter

SSC

Side scatter

BBB

Blood-brain barrier

Authors’ contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Minghang Liu, Kai Zhao and Chenxuan Yang. The first draft of the manuscript was written by Minghang Liu, Guo chen Sun and Dongdong Wu.Supervision and final revision were performed by Yanlian Yang and Bainan Xu. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The datasets generated during and analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the First Medical Center of the Chinese PLA General Hospital (Approval No. S2018-268-02).Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.