Skip to main content
NCBI home page
Springer logoLink to Springer
. 2026 May 5;28(10):4892–4902. doi: 10.1007/s12094-026-04336-2

Expression and diagnostic significance of plasma exosomal RNA in diffuse large B-cell lymphoma

Xiaomei Ma 1,#, Zimiao Zhang 1,#, Jiayi Deng 1, Yueyuan Lai 1, Aili Zhang 1, Congjie Chen 1, Xiaohui Shangguan 1, Weihao Wu 1,✉, Longtian Chen 1,✉
PMCID: PMC13601143  PMID: 42084824

Abstract

Objective

This study aimed to investigate the expression profiles of plasma exosomal RNAs in patients with Diffuse Large B-Cell Lymphoma (DLBCL) and evaluate their diagnostic potential.

Methods

Exosomes were isolated from 9 DLBCL patients and 9 healthy controls, followed by whole transcriptome RNA sequencing to identify differentially expressed RNAs. Key RNAs were selected and validated via qRT-PCR using an independent cohort of 41 treatment-naïve DLBCL patients and 41 healthy controls. The StarBase database was utilized to predict target relationships and construct lncRNA/miRNA competing endogenous RNA (ceRNA) pairs. Diagnostic efficacy was assessed using Receiver Operating Characteristic (ROC) curve analysis.

Results

High-throughput sequencing revealed significant RNA expression differences between DLBCL patients and healthy controls. Specifically, 35 microRNAs (miRNAs), 1604 long non-coding RNAs (lncRNAs), and 330 messenger RNAs (mRNAs) were differentially expressed in the DLBCL group, while no significant differences were observed in circular RNA (circRNA) expression. Two ceRNA pairs, MALAT1/hsa-miR-181a-5p and SLC9A3-AS1/hsa-miR-10a-5p, were constructed. qRT-PCR validation confirmed that hsa-miR-181a-5p and hsa-miR-10a-5p were significantly down-regulated, while SLC9A3-AS1 was up-regulated in the DLBCL group; MALAT1 expression showed no significant difference. ROC analysis demonstrated Area Under the Curve (AUC) values of 0.813 for hsa-miR-181a-5p, 0.800 for hsa-miR-10a-5p, and 0.787 for SLC9A3-AS1. The combined AUC for SLC9A3-AS1 and hsa-miR-10a-5p was 0.8226, and for SLC9A3-AS1 and hsa-miR-181a-5p was 0.852.

Conclusion

Exosome-derived miRNAs, lncRNAs, and mRNAs exhibit distinct expression patterns in DLBCL patients compared to healthy controls. The RNAs hsa-miR-181a-5p, hsa-miR-10a-5p, and SLC9A3-AS1 show promise as diagnostic biomarkers for DLBCL, with combined RNA panels offering improved diagnostic efficacy over single markers. The SLC9A3-AS1/hsa-miR-10a-5p pair may form a regulatory network influencing DLBCL pathogenesis.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12094-026-04336-2.

Keywords: Diffuse large B-cell lymphoma, Exosome, RNA, Biomarker

Introduction

Diffuse large B-cell lymphoma (DLBCL) is the most prevalent subtype of non-Hodgkin lymphoma, accounting for 30–40% of cases worldwide [1]. With the aging global population, DLBCL incidence is rising annually and is generally higher in males than females [2]. The pathogenesis of DLBCL is complex, involving genetic abnormalities, epigenetic alterations, and interactions with the tumor microenvironment, and can arise from various stages of B-cell development [3, 4]. DLBCL is highly aggressive and heterogeneous, displaying significant biological, pathological, and clinical variability among patients [5]. Early-stage DLBCL often presents with non-specific symptoms such as painless lymphadenopathy, fever, night sweats, or unexplained weight loss. Currently, no circulating biomarkers with diagnostic utility have been established, and diagnosis typically relies on excisional biopsy of affected tissues following significant lymph node enlargement or extranodal involvement [6]. While histopathological biopsy remains the gold standard, it has limitations in sampling deep-seated lesions, dynamic monitoring, and guiding personalized therapy. Tumor heterogeneity further complicates the representativeness of single biopsies, and the invasive nature of repeated procedures may increase complication risks and reduce patient compliance. Thus, there is an urgent need for minimally invasive, sensitive, and specific biomarkers to aid in diagnosis and disease monitoring.

Liquid biopsy has emerged as a promising approach due to its non-invasiveness, ease of specimen acquisition, and potential for dynamic monitoring. By analyzing components such as circulating tumor cells, circulating tumor DNA (ctDNA), or exosomes in blood and other body fluids, it is possible to capture real-time molecular characteristics and dynamic changes of tumors [7]. This provides new avenues for DLBCL diagnosis and monitoring, offering dynamic insights for precision medicine.

Exosomes are lipid bilayer vesicles released by cells [8], present in various body fluids, and derived from both normal and tumor cells. They carry bioactive molecules from their parent cells, including proteins, lipids, DNA, and RNA [9]. Accumulating evidence indicates that tumor-derived exosomes participate in key pathological processes such as tumor microenvironment remodeling, immune evasion, inflammation, and angiogenesis [10–12], thereby promoting tumorigenesis and progression. They hold clinical potential as novel biomarkers for early cancer detection, prognosis assessment, and treatment monitoring. This study aims to systematically analyze plasma exosomal RNA expression profiles in treatment-naïve DLBCL patients and healthy controls using whole transcriptome sequencing, identify differentially expressed RNAs, construct a ceRNA network, and preliminarily screen molecular biomarkers for DLBCL diagnosis via ROC analysis, thereby providing new insights for clinical management.

Materials and methods

Study subjects

Peripheral venous blood samples were collected from 50 newly diagnosed, untreated DLBCL patients (experimental group) and 50 healthy volunteers (control group) at Longyan First Hospital Affiliated to Fujian Medical University between October 2022 and October 2024. Detailed clinical data, including gender, age, Hans classification, Ann Arbor stage, International Prognostic Index (IPI) score, and lactate dehydrogenase (LDH) levels, were recorded. The study was approved by the hospital's Ethics Committee. All DLBCL patients met the diagnostic criteria of the 2017 WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues [13], with diagnoses confirmed by histopathological biopsy reviewed by two senior pathologists. Healthy controls were recruited from the same hospital's health examination center. Subjects with infections, other malignancies, severe systemic diseases, hematological disorders, autoimmune diseases, or incomplete clinical records were excluded.

Experimental materials and reagents

FITC Mouse Anti-Human CD63 and FITC Mouse Anti-Human CD81 were purchased from BD Biosciences.A transmission electron microscope and ultracentrifuge were from Hitachi. The nanoparticle size analyzer was from NanoFCM. The high-speed benchtop refrigerated centrifuge was from Hunan Xiangyi Laboratory Instrument Development Co., Ltd. The qPCR instrument was from Bio-Rad.

Experimental methods

Sample collection

Peripheral blood was collected from fasting subjects in the morning into EDTA anticoagulant tubes and centrifuged within 1 h. The plasma layer was carefully transferred into cryovials and stored at − 80 °C. Hemolysis was avoided throughout the process.

Plasma exosome isolation

Plasma was separated from venous blood collected from the elbow. Exosomes were extracted using ultracentrifugation. Centrifuge the plasma sample at 2000g at 4 °C for 30 min. Then transfer the supernatant to a new centrifuge tube and centrifuge at 10,000g at 4 °C for 45 min to remove larger vesicles. Take the supernatant, filter it through a 0.45 μm filter membrane, and collect the filtrate. Centrifuge it at 4 °C at 100,000g for 70 min. Remove the supernatant, resuspend with 10 mL of pre-cooled 1×PBS, select an ultracentrifuter, filter through a 0.22 μm filter membrane, and then centrifuge at 4 °C, 100,000g, for 70 min. Remove the supernatant and resuspend with 400 μL of pre-cooled 1×PBS. 20 μL of the isolated exosome samples were taken for electron microscopy, 10 μL for particle size, 20 μL for fluorescence, and the remaining exosome samples were stored in a – 80 °C refrigerator.

Transmission electron microscopy (TEM)

A 10 μL exosome sample was applied to a copper grid,adsorbed for 1 min, and excess liquid removed. Then, 10 μL phosphotungstic acid was added, incubated for 1 min, and blotted. After air-drying, exosome morphology was observed and imaged using TEM.

Nanoparticle tracking analysis (NTA)

A 10 μL exosome sample was diluted to 30 μL. Instrument performance was verified using a standard prior to sample loading. Gradient dilution was applied to prevent needle clogging. Particle size and concentration data were obtained using the NanoFCM instrument.

Fluorescence labeling and nano-flow cytometry

A 20 μL exosome sample was diluted to 60 μL. Then, 30 μL of diluted exosomes was mixed with 20 μL of fluorescently labeled antibody (CD63, CD81), incubated at 37 °C in the dark for 30 min. After adding 1 mL of pre-cooled PBS, the mixture was ultracentrifuged at 110,000g for 70 min at 4 °C. The supernatant was discarded, and the pellet was resuspended in 50 μL of pre-cooled PBS. Instrument performance was checked before sample loading. Gradient dilution was used to prevent clogging. Protein marker results were obtained post-detection.

Exosomal RNA high-throughput sequencing

Referring to the relevant research on the exploratory study of exosome RNA sequencing, every 3 plasma samples were mixed into a uniform sample and three biological replicates were conducted. A total of 9 samples were required. Therefore, we randomly selected 9 samples from each of the experimental group and the control group, total RNA was isolated from characterized plasma exosomes. RNA concentration was measured using a Quantus Fluorometer. Whole transcriptome and miRNA-specific libraries were constructed using TruSeq Small RNA Sample Prep Kits (Illumina). Libraries underwent quality control on the Agilent 2100 bioanalyzer. All samples had RIN > 7.0 and were sequenced on the Illumina HiSeq 2000/2500 platform (full transcriptome matching pairs 150 base pairs;miRNA single ends 50 base pairs). Raw data were filtered to obtain clean data for bioinformatics analysis. Differential expression analysis was performed using DESeq R package (v3.0.3) with criteria: |log2FC|>1 and p-value <0.05. We did apply the False Detection Rate (FDR) correction in the differential expression screening. After correction, Q value < 0.05 was set as one of the significance thresholds.

ceRNA network construction

Differentially expressed miRNAs and lncRNAs were screened.The StarBase database was used to predict miRNA-lncRNA target relationships with default parameters (Clade: Mammal, Genome: Human, Assembly: hg38, Supporting experiments: ≥3, Pan-cancer≥0).

qRT-PCR

To validate sequencing data, qRT-PCR was performed on exosome samples from an additional 41 DLBCL patients and 41 healthy controls to verify the expression of candidate RNAs. qRT-PCR uses GAPDH as an internal reference.

Dissolve RNA template, 5×Reaction Buffer, 10 mM dNTP mix, RI, RT and DEPC Water and place them on ice to prepare the reaction system. After incubation at 65 °C for 5 min, add the above Mix, vortex and shake, and briefly centrifuge. Incubate at 25 °C for 5 min, at 42 °C for 60 min, and at 70 °C for 5 min. After the reaction, cool on ice and store for a long time at – 80 °C. The reverse transcription of mRNA and lncRNA uses oligo dT and random primers. The miRNA and internal reference U6 reverse transcription primer sequences are shown in Supplementary Table 1. The primer sequences of qRT-PCR are shown in Supplementary Table 2.

Statistical analysis

Data were analyzed using SPSS (v27.0). Graphs were generated with GraphPad Prism (v9.0) and Cytoscape (v3.10.0). Normally distributed data are presented as mean±SD and compared using independent samples t-test. Non-normally distributed data are expressed as median (interquartile range) and compared using non-parametric tests. Categorical data are presented as rates/proportions and compared using chi-square test. A p-value<0.05 was considered statistically significant. The diagnostic value of differentially expressed RNAs was evaluated by calculating the Area Under the Curve (AUC) via ROC analysis.

The performance analysis and calculation using the PASS software show that with α = 0.05 and P = 0.7, the estimated minimum sample size required is approximately 41 cases per group.

Results

General clinical characteristics

Plasma samples were collected from 50 DLBCL patients and 50 healthy controls. Among the patients, 23 (46%) were male and 27 (54%) female, with a mean age of 63.7±12.9 years; 31 (62%) were over 60 years old. The healthy control group consisted of 24 males (48%) and 26 females (52%), with a mean age of 59.1 ± 11.5 years. No statistically significant differences were observed between the two groups in terms of age or gender. According to Hans classification, 17 (34%) were germinal center B-cell-like (GCB), 29 (58%) were non-GCB, and 4 (8%) were unclassifiable. By Ann Arbor staging, 16 were stages I–II, 32 were stages III–IV, and 1 (2%) was unstageable. Based on IPI, 27 (54%) were low-intermediate risk (scores 0–2), 20 (40%) were intermediate-high risk (scores 3–5), and 3 (6%) were unscoreable. LDH levels were elevated in 27 patients (54%) and normal in 23 (46%) (Table 1).

Table 1.

Clinical data characteristics of DLBCL patients and healthy volunteers

Variable DLBCL patients (N = 50) Healthy volunteers (N = 50) P
Gender (n, %)
 Male 23 (46.0%) 24 (48.0%) 0.841
 Female 27 (54.0%) 26 (52.0%)
Age (years)
  ≤ 60 (n, %) 19 (38.0%) 27 (54.0%) 0.064
  > 60 (n, %) 31 (62.0%) 23 (46.0%)
Subtype (n, %) NA
 GCB 17 (34.0%)
 Non-GCB 29 (58.0%)
 Non-subtype 4 (8.0%)
Ann Arbor stage NA
 Ⅰ–Ⅱ 16 (32.0%)
 Ⅲ–Ⅳ 33 (66.0%)
 Non- stage 1 (5.0%)
IPI score NA
 Low-to medium-risk group (0–2 score) 27 (54.0%)
 High-to medium-risk group (3–5 score) 20 (40.0%)
 Non-score 3 (6.0%)
LDH(IU/L) 257.50 (212.00424.25) NA
  > 247 27 (54.0%)
  ≤ 247 23 (46.0%)
Open in a new tab

DLBCL diffuse large B cell lymphoma, GCB germinal center B cell subtype, IPI international prognostic index, LDH lactate dehydrogenase

Identification of plasma exosomes

Exosomes isolated from both groups exhibited a round or oval shape with a typical double-membrane structure under TEM (Fig. 1A). NTA showed a uniform particle size distribution with a single-peak curve. The average exosome diameter was 70.3 nm in the experimental group and 84.8 nm in the control group (Fig. 1B), within the expected exosome size range. NanoFCM confirmed the presence of CD63 and CD81 surface markers in all samples (Figs. 1C, D), the positive expression rates of exosome surface proteins CD63 and CD81 in the DLBCL patient group were 2.1% and 6.3% respectively, while those in the healthy control group were 13.1% and 13.6% respectively, verifying exosome quality and suitability for RNA sequencing.

Fig. 1.

Fig. 1

Open in a new tab

Morphology and characterization of exosomes. A The morphology of exosomes was observed by transmission electron microscopy (TEM), showing typical exosomes with round or oval, bilayer structure, B the particle size distribution of exosomes showed unimodal normal distribution, C detection of exosom-specific signature proteins CD63 and CD81 in the experimental group, D detection of exosom-specific signature proteins CD63 and CD81 in the negative control group

Analysis of differentially expressed RNAs in plasma exosomes

Whole transcriptome sequencing of plasma exosomes from 9 DLBCL patients and 9 healthy controls identified significant differences:35 miRNAs (5 up, 30 down) (Figs. 2A, 3A), 1604 lncRNAs (826 up, 778 down) (Figs. 2B, 3B), and 330 mRNAs (137 up, 193 down) (Fig. 2C). No significant differences in circRNA expression were observed (Fig. 2D).

Fig. 2.

Fig. 2

Open in a new tab

Molecular volcano maps showing differentially expressed RNAs in plasma exosomes between DLBCL patients (n = 9) and healthy controls (n = 9) by whole transcriptome sequencing. Differential expression criteria: |log2FC|> 1 and FDR-adjusted P-value < 0.05 (A miRNA, B lncRNA, C mRNA, D circRNA)

Fig. 3.

Fig. 3

Open in a new tab

Clustering heat map of differential miRNA and lncRNA expression in plasma exosomes from DLBCL patients (n = 9) and healthy controls (n = 9)

Construction of ceRNA pairs

Based on sequencing results and literature,two abundantly and differentially expressed miRNAs, hsa-miR-181a-5p and hsa-miR-10a-5p, were selected. StarBase analysis identified 37 lncRNAs targeting hsa-miR-181a-5p and 18 targeting hsa-miR-10a-5p (Fig. 4). Integrating CLIP data and recent research, two lncRNAs, MALAT1 and SLC9A3-AS1, were chosen for further study.

Fig. 4.

Fig. 4

Open in a new tab

Lncrnas targeting hsa-miR-181a-5p and hsa-miR-10a-5p (StarBase database was used to evaluate miRNA-lncRNA interactions, network mapping was performed by cytoscape software)

Expression levels of ceRNA pairs in DLBCL patients

qRT-PCR validation in 41 DLBCL patients and 41 controls revealed significant differences in exosomal levels of SLC9A3-AS1, hsa-miR-181a-5p, and hsa-miR-10a-5p. SLC9A3-AS1 was significantly up-regulated (P<0.0001), while hsa-miR-181a-5p (P<0.05) and hsa-miR-10a-5p (P<0.0001) were down-regulated in the DLBCL group (Fig. 5). MALAT1 expression showed no significant difference.

Fig. 5.

Fig. 5

Open in a new tab

qRT-PCR results (qRT-PCR assay was used to detect the expression levels of hsa-miR-181a-5p, hsa-miR-10a-5p, MALAT1 and SLC9A3-AS1 in the patient group (n = 41) and the healthy group (n = 41), with GAPDH as the internal reference, calculated by 2-△△Ct method. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Diagnostic value of differentially expressed exosomal RNAs

ROC analysis yielded AUC values of 0.787 for SLC9A3-AS1 (sensitivity 68.3%, specificity 85.4%; Fig. 6A), 0.800 for hsa-miR-10a-5p (sensitivity 58.5%, specificity 90.2%; Fig. 6B), and 0.813 for hsa-miR-181a-5p (sensitivity 78%, specificity 78%; Fig. 6C). The combined AUC for SLC9A3-AS1 and hsa-miR-181a-5p was 0.852 (sensitivity 80.5%, specificity 82.9%; Fig. 6D), and for SLC9A3-AS1 and hsa-miR-10a-5p was 0.826 (sensitivity 68.3%, specificity 90.2%; Fig. 6E).

Fig. 6.

Fig. 6

Open in a new tab

ROC curve analysis of the diagnostic value of hsa-miR-181a-5p, hsa-miR-10a-5p, SLC9A3-AS1 and SLC9A3-AS1 combined with hsa-miR-181a-5p and hsa-miR-10a-5p in the patient group (n = 41) and the healthy group (n = 41). A Evaluation of the diagnostic value of SLC9A3-AS1 in DLBCL, B evaluation of diagnostic value of hsa-miR-10a-5p in DLBCL, C evaluation of diagnostic value of hsa-miR-181a-5p in DLBCL, D evaluation of the diagnostic value of SLC9A3-AS1 combined with hsa-miR-181a-5p in DLBCL, E evaluation of the diagnostic value of SLC9A3-AS1 combined with hsa-miR-10a-5p in DLBCL.( AUC calculation: Area under the curve was computed using the trapezoidal rule; combined ROC analysis was performed using binary logistic regression predicted probabilities)

Discussion

Exosomes derived from DLBCL cells exhibit distinct compositional profiles at the RNA level compared to those from healthy individuals. Previous studies highlight the role of exosomal miRNAs in DLBCL diagnosis, prognosis, and treatment response [14]. For instance, miRNAs like miR-155, miR-21, and miR-451a show considerable diagnostic and prognostic value, and their combined detection may improve accuracy and aid in distinguishing DLBCL from other lymphomas [15–17]. However, current research often focuses on single RNA types, lacking systematic investigation into inter-RNA interactions and regulatory networks. Challenges such as insufficient biomarker specificity, non-standardized detection methods, and limited large-scale validation remain. Therefore, identifying and validating sensitive and specific exosomal RNA markers, constructing their regulatory networks, and exploring their diagnostic utility in DLBCL are of significant scientific and clinical importance.

This study employed high-throughput whole transcriptome sequencing to comprehensively analyze exosomal RNA expression in DLBCL. Results revealed significant differential expression of miRNAs, lncRNAs, and mRNAs, but not circRNAs, suggesting important roles for the former in DLBCL pathogenesis. Among these, hsa-miR-181a-5p and hsa-miR-10a-5p showed prominent differential expression.

miRNAs are short non-coding RNAs (~20–24 nt) involved in B-cell development and frequently dysregulated in B-cell lymphomas [18, 19], making them potential DLBCL biomarkers [20–22]. They can act as tumor suppressors or oncogenes. For example, miR-181a inhibits tumor cell proliferation via AKT, MAPK, RAS, and mTOR pathways [23]. Kozloski et al. demonstrated that miR-181a overexpression suppresses NF-κB signaling by targeting REL, CARD11, RELA, NFKB1A, and NFKB1, reducing tumor cell survival and proliferation [24]. Similarly, miR-10a-5p is dysregulated in various malignancies. In DLBCL, it inhibits cell proliferation and promotes apoptosis by negatively regulating BCL6, and its decreased expression may drive malignant transformation [25]. Our qRT-PCR results confirmed the significant down-regulation of hsa-miR-181a-5p and hsa-miR-10a-5p, consistent with sequencing data, supporting their potential as auxiliary diagnostic markers.

Salmena et al. [26] proposed the ceRNA hypothesis, suggesting that mRNAs and non-coding RNAs form interconnected regulatory networks. This implies that RNA molecules participate in extensive post-transcriptional regulation during physiological homeostasis and pathological processes, highlighting ceRNAs' potential as diagnostic markers or therapeutic targets [27]. Within ceRNA networks, lncRNAs act as miRNA sponges to regulate mRNA expression. LncRNAs (>200 nt) do not encode proteins but regulate gene expression at multiple levels [28, 29]. To explore whether lncRNAs upstream of miRNAs contribute to DLBCL progression, we used StarBase to predict targets for hsa-miR-181a-5p and hsa-miR-10a-5p, identifying 37 and 18 lncRNAs, respectively. SLC9A3-AS1 and MALAT1 have documented roles in hematological tumors. MALAT1 is overexpressed in multiple myeloma, DLBCL, and chronic lymphocytic leukemia [30], potentially sponging miR-195 to regulate PD-L1 and modulate DLBCL cell proliferation [31]. In acute myeloid leukemia, MALAT1 influences CXCR4 via miR-146a, affecting cell proliferation, migration, and apoptosis [32]. SLC9A3-AS1, which modulates colon cancer cell activity by negatively regulating miR-486, is a potential non-invasive diagnostic marker for colon cancer [33]. High SLC9A3-AS1 levels correlate with poor prognosis in various cancers [34–36]. In our study, qRT-PCR confirmed significant up-regulation of SLC9A3-AS1 in DLBCL patient exosomes. Its negative correlation with miR-10a-5p suggests a potential ceRNA regulatory axis in DLBCL progression. Interestingly, MALAT1 showed no significant differential expression, possibly due to DLBCL subtypes, sample size, or heterogeneity, warranting further validation.

Nevertheless, it is important to acknowledge that the proposed ceRNA interactions (SLC9A3-AS1/hsa-miR-10a-5p and MALAT1/hsa-miR-181a-5p) in this study are primarily derived from bioinformatics predictions based on the StarBase database and the observed inverse expression correlations. While these findings provide a hypothesis-generating framework, they do not constitute direct evidence of a functional ceRNA network. Experimental validation is essential to establish causality. Future studies should employ dual-luciferase reporter assays to confirm the direct binding between these lncRNAs and miRNAs, and perform functional assays (e.g., miRNA mimics/inhibitors, siRNA-mediated knockdown) in DLBCL cell lines to elucidate the biological consequences of disrupting these putative interactions. Such investigations will be crucial to determine whether the SLC9A3-AS1/miR-10a-5p axis genuinely contributes to DLBCL pathogenesis and to assess its potential as a therapeutic target. Our cohort is mainly composed of non-GCB subtype DLBCL, and the expression level of MALAT1 may vary depending on the molecular subtype. MALAT1 dysregulation is more obvious in specific DLBCL subgroups. Secondly, the abundance of circulating MALAT1 in plasma may be relatively low, and qRT-PCR detection may be affected by pre-analysis variables, RNA integrity or detection sensitivity. Future research should also explore whether the expression of MALAT1 in exosomes is associated with specific clinical subpopulations or disease stages.

ROC analysis systematically evaluated the diagnostic potential of hsa-miR-181a-5p, hsa-miR-10a-5p, and SLC9A3-AS1. All individual AUCs exceeded 0.7, indicating diagnostic value. Combined RNA panels (SLC9A3-AS1 with either miRNA) yielded higher AUCs than single markers, suggesting improved diagnostic accuracy, consistent with the multi-molecular synergy concept in tumor heterogeneity. These findings support the potential of constructing a multi-marker exosomal RNA model for DLBCL diagnosis.

In this study, the RNA map of DLBCL exosomes was mapped through the whole transcriptome sequencing system. The SLC9A3-AS1/miR-10a-5p ceRNA regulatory axis was constructed for the first time, and the diagnostic advantages of multi-RNA combined detection were confirmed, providing a new strategy for liquid biopsy of DLBCL. Plasma exosomal RNA expression profiles are significantly altered in DLBCL. In the SLC9A3-AS1/hsa-miR-10a-5p pair, miR-10a-5p was down-regulated while SLC9A3-AS1 was up-regulated, suggesting a potential interactive role in DLBCL pathogenesis, though the specific mechanism requires further investigation. This study preliminarily identifies hsa-miR-181a-5p, hsa-miR-10a-5p, and SLC9A3-AS1 as candidate biomarkers for DLBCL diagnosis. The diagnostic value assessment of candidate exosome RNA markers in this study is still in its infancy. As a single-center study with a limited sample size, we were unable to directly compare RNA with LDH, IPI scores, subtype etc., nor did we systematically analyze its correlation with clinicopathological features. This type of analysis is crucial for evaluating the incremental diagnostic value and clinical practicality of new markers, but it needs to be conducted in larger-scale multi-center cohorts to avoid insufficient statistical power and bias caused by small subgroup sample sizes. In the future, we will expand the sample size and carry out multi-center prospective studies to verify the expression differences of these RNA markers in different clinical subgroups. And a multi-factor regression model was utilized to evaluate its incremental value based on traditional clinical parameters. Meanwhile, we will explore the correlation between RNA and clinical features, analyze their role in the occurrence and development of DLBCL, and provide a basis for individualized diagnosis and treatment.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 13 KB) (13.5KB, docx)

Acknowledgements

The authors thank all the participants of the study.

Author contributions

MXM, ZMZ, and LTC conceptualized and designed the study, with LTC and MXM serving as guarantors. ZMZ and JYD conducted sample collection, while YYL and WHW coordinated all data acquisition. ALZ and CJC performed systematic analysis and interpretation of the data. XHSG designed and plotted the figures and tables. MXM, and ZMZ wrote the original manuscript draft. LTC, CJC, and WHW critically reviewed and edited the final version. This work was funded by grants from LTC and XMMC. All authors (XMM, ZMZ, JYD, YYL, ALZ, CJC, XHSG, WHW, LTC) approved the submitted manuscript.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported by the Fujian Provincial Natural Science Foundation Program [grant number 2023J011894], the Longyan Municipal Science and Technology Planning Project [grant number 2022LYF17110].

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. The datasets used in this study can be obtained from the SRA database. The accession number is PRJNA1433574.

Declarations

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Ethical approval

This study was approved by the Ethics Committee of Longyan First Hospital (Approval No. LYREC2024-k172-01). Whole blood samples from 50 DLBCL patients and 50 healthy controls were collected after obtaining written informed consent. All procedures followed the Declaration of Helsinki.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Xiaomei Ma and Zimiao Zhang contributed equally to this work.

Contributor Information

Weihao Wu, Email: weihao0077@sina.com.

Longtian Chen, Email: ltchen@163.com.

References

  • 1.Li S, Young KH, Medeiros LJ. Diffuse large B-cell lymphoma. Pathology. 2018;50(1):74–87. [DOI] [PubMed] [Google Scholar]
  • 2.Wang SS. Epidemiology and etiology of diffuse large B-cell lymphoma. Semin Hematol. 2023;60(5):255–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Swerdlow SH, Campo E, Pileri SA, Harris NL, Stein H, Siebert R, et al. The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood. 2016;127(20):2375–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Barraclough A, Hawkes E, Sehn LH, Smith SM. Diffuse large B-cell lymphoma. Hematol Oncol. 2024;42(6):e3202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Susanibar-Adaniya S, Barta SK. 2021 update on diffuse large B cell lymphoma: a review of current data and potential applications on risk stratification and management. Am J Hematol. 2021;96(5):617–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Alaggio R, Amador C, Anagnostopoulos I, Attygalle AD, Araujo IBO, Berti E, et al. Correction: “the 5th edition of the World Health Organization classification of Haematolymphoid tumours: Lymphoid Neoplasms.” Leukemia. 2022;36(7):1720–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Lone SN, Nisar S, Masoodi T, Singh M, Rizwan A, Hashem S,et al. Liquid biopsy: a step closer to transform diagnosis, prognosis and future of cancer treatments. Mol Cancer. 2022;21(1):79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Rayamajhi S, Aryal S. Surface functionalization strategies of extracellular vesicles. J Mater Chem B. 2020;8(21):4552–69. [DOI] [PubMed] [Google Scholar]
  • 9.Liang Y, Duan L, Lu J, Xia J. Engineering exosomes for targeted drug delivery. Theranostics. 2021;11(7):3183–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Liu J, Wang X. Focus on exosomes-from pathogenic mechanisms to the potential clinical application value in lymphoma. J Cell Biochem. 2019;120(12):19220–8. [DOI] [PubMed] [Google Scholar]
  • 11.Mashouri L, Yousefi H, Aref AR, Ahadi AM, Molaei F, Alahari SK. Exosomes: composition, biogenesis, and mechanisms in cancer metastasis and drug resistance. Mol Cancer. 2019;18(1):75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Paskeh M, Entezari M, Mirzaei S, Zabolian A, Saleki H, Naghdi MJ,et al. Emerging role of exosomes in cancer progression and tumor microenvironment remodeling. J Hematol Oncol. 2022;15(1):83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Polyatskin IL, Artemyeva AS, Krivolapov YA. [Revised WHO classification of tumors of hematopoietic and lymphoid tissues, 2017 (4th edition): lymphoid tumors]. Arkh Patol. 2019;81(3):59–65. [DOI] [PubMed] [Google Scholar]
  • 14.Lopez-Santillan M, Larrabeiti-Etxebarria A, Arzuaga-Mendez J, Lopez-Lopez E, Garcia-Orad A. Circulating miRNAs as biomarkers in diffuse large B-cell lymphoma: a systematic review. Oncotarget. 2018;9(32):22850–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Musilova K, Mraz M. MicroRNAs in B-cell lymphomas: how a complex biology gets more complex. Leukemia. 2015;29(5):1004–17. [DOI] [PubMed] [Google Scholar]
  • 16.Larrabeiti-Etxebarria A, Lopez-Santillan M, Santos-Zorrozua B, Lopez-Lopez E, Garcia-Orad A. Systematic review of the potential of microRNAs in diffuse large B cell lymphoma. Cancers (Basel). 2019. 10.3390/cancers11020144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Cao D, Cao X, Jiang Y, Xu J, Zheng Y, Kang D, et al. Circulating exosomal microRNAs as diagnostic and prognostic biomarkers in patients with diffuse large B-cell lymphoma. Hematol Oncol. 2022;40(2):172–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Liu B, Li J, Cairns MJ. Identifying miRNAs, targets and functions. Brief Bioinform. 2014;15(1):1–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Sole C, Larrea E, Di Pinto G, Tellaetxe M, Lawrie CH. miRNAs in B-cell lymphoma: Molecular mechanisms and biomarker potential. Cancer Lett. 2017;405:79–89. [DOI] [PubMed] [Google Scholar]
  • 20.Alsaadi M, Khan MY, Dalhat MH, Bahashwan S, Khan MU, Albar A, et al. Dysregulation of miRNAs in DLBCL: causative factor for pathogenesis, diagnosis and prognosis. Diagnostics Basel. 2021. 10.3390/diagnostics11101739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bouvy C, Wannez A, George F, Graux C, Chatelain C, Dogne J. Circulating MicroRNAs as biomarkers in Diffuse Large B-cell Lymphoma: a pilot prospective longitudinal clinical study. Biomark Cancer. 2018;10:1179299X-18781095X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Larrabeiti-Etxebarria A, Bilbao-Aldaiturriaga N, Arzuaga-Mendez J, Martin-Arruti M, Cozzuto L, Gaafar A, et al. microRNA sequencing for biomarker detection in the diagnosis, classification and prognosis of Diffuse Large B Cell Lymphoma. Sci Rep. 2023;13(1):12159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Amiri BS, Sabernia N, Abouali B, Amini P, Rezaeeyan H. Evaluation of MicroRNA as minimal residual disease in leukemia: diagnostic and prognostic approach: a review. Iran J Public Health. 2023;52(12):2541–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kozloski GA, Jiang X, Bhatt S, Ruiz J, Vega F, Shaknovich R, et al. miR-181a negatively regulates NF-kappaB signaling and affects activated B-cell-like diffuse large B-cell lymphoma pathogenesis. Blood. 2016;127(23):2856–66. [DOI] [PubMed] [Google Scholar]
  • 25.Fan Q, Meng X, Liang H, Zhang H, Liu X, Li L, et al. miR-10a inhibits cell proliferation and promotes cell apoptosis by targeting BCL6 in diffuse large B-cell lymphoma. Protein Cell. 2016;7(12):899–912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP. A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language? Cell. 2011;146(3):353–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Qi X, Zhang DH, Wu N, Xiao JH, Wang X, Ma W. ceRNA in cancer: possible functions and clinical implications. J Med Genet. 2015;52(10):710–8. [DOI] [PubMed] [Google Scholar]
  • 28.Chi Y, Wang D, Wang J, Yu W, Yang J. Long non-coding RNA in the pathogenesis of cancers. Cells. 2019. 10.3390/cells8091015. [DOI] [Google Scholar]
  • 29.Zhang Y, Mao Q, Xia Q, Cheng J, Huang Z, Li Y, et al. Noncoding RNAs link metabolic reprogramming to immune microenvironment in cancers. J Hematol Oncol. 2021;14(1):169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Zhang C, Qin Y, Wu Y, Xu Y, Xu H, Shu Y. Long non-coding RNA MALAT1 in hematological malignancies and its clinical applications. Chin Med J (Engl). 2024;137(10):1151–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Wang QM, Lian GY, Song Y, Huang YF, Gong Y. LncRNA MALAT1 promotes tumorigenesis and immune escape of diffuse large B cell lymphoma by sponging miR-195. Life Sci. 2019;231:116335. [DOI] [PubMed] [Google Scholar]
  • 32.Sheng XF, Hong LL, Li H, Huang FY, Wen Q, Zhuang HF. Long non-coding RNA MALAT1 modulate cell migration, proliferation and apoptosis by sponging microRNA-146a to regulate CXCR4 expression in acute myeloid leukemia. Hematology. 2021;26(1):43–52. [DOI] [PubMed] [Google Scholar]
  • 33.Xu X, Qing H, Jiang C, Zhao X, Wei J. Influence of the lncRNA SLC9A3-AS1 on colon cancer and the biological activities of colon cancer cells. Discov Oncol. 2025;16(1):358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Oduor CI, Movassagh M, Kaymaz Y, Chelimo K, Otieno J, Ongecha JM, et al. Human and Epstein-Barr Virus miRNA Profiling as Predictive Biomarkers for Endemic Burkitt Lymphoma. Front Microbiol. 2017;8:501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Gao F, Wu Q, Lu D. MicroRNA-10a-5p-mediated downregulation of GATA6 inhibits tumor progression in ovarian cancer. Hum Cell. 2024;37(1):271–84. [DOI] [PubMed] [Google Scholar]
  • 36.Gu Y, Feng X, Jin Y, Liu Y, Zeng L, Zhou D, et al. Upregulation of miRNA-10a-5p promotes tumor progression in cervical cancer by suppressing UBE2I signaling. J Obstet Gynaecol. 2023;43(1):2171283. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file1 (DOCX 13 KB) (13.5KB, docx)

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. The datasets used in this study can be obtained from the SRA database. The accession number is PRJNA1433574.

Articles from Clinical & Translational Oncology are provided here courtesy of Springer

RESOURCES