ENO1, a Key Glycolytic Enzyme, Predicts Prognosis and Correlates with Immune Microenvironment in Primary Gastrointestinal Diffuse Large B-Cell Lymphoma

Abstract

Background: Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma, with the gastrointestinal tract being the predominant extranodal site. Primary gastrointestinal DLBCL (PGI-DLBCL) arises in a chronic inflammatory microenvironment that promotes aerobic glycolysis. Alpha-enolase (ENO1), a key glycolytic enzyme, is overexpressed in various malignancies and correlates with poor prognosis. However, its prognostic role and relationship with the immune microenvironment in PGI-DLBCL remain unclear. This study explored the prognostic significance of ENO1 in PGI-DLBCL and its associations with tumor-infiltrating lymphocytes (TILs), tumor-associated macrophages (TAMs) polarization, and lymphocyte-to-monocyte ratio (LMR).

Materials and methods: A retrospective cohort of 95 newly diagnosed PGI-DLBCL patients was enrolled. LMR was calculated. Immunohistochemistry was performed to detect ENO1 expression and infiltration of CD4⁺, CD8⁺, Foxp3⁺ T cells, and CD68⁺ macrophages. Multiplex immunofluorescence assessed M1/M2 macrophage polarization. Correlations between ENO1, immune infiltrates, LMR, and clinicopathological features were analyzed. Survival outcomes were assessed using methods such as Kaplan–Meier analysis and Cox regression. The study was approved by the institutional ethics committee.

Results: High ENO1 expression was correlated with increased macrophage infiltration and M2 polarization (elevated M2/total macrophage and M2/M1 ratios). ENO1 expression was also correlated with LMR, and low LMR was associated with reduced CD8⁺ T-cell infiltration. Multivariate analysis identified LMR and CD4⁺ T-cell infiltration as protective factors, whereas high ENO1 expression emerged as an adverse prognostic factor.

Discussion: This study suggests that ENO1 may serve as an independent prognostic biomarker in PGI-DLBCL. Its expression level correlates with LMR and may be involved in macrophage recruitment and polarization toward the M2 phenotype. There may be an intrinsic link among local immunity, circulating immunity, and metabolism, although the underlying mechanisms remain unclear and warrant further investigation.

Conclusion: ENO1 is a prognostic biomarker in PGI-DLBCL, positively linked to M2 macrophage polarization and associated with systemic immune status reflected by LMR.

Graphical abstract

Keywords

alpha-enolase / primary gastrointestinal diffuse large B-Cell lymphoma / tumor microenvironment

Introduction

Diffuse large B-cell lymphoma (DLBCL) is the most common and highly aggressive/heterogeneous subtype of non-Hodgkin lymphoma[1]. The gastrointestinal tract is the most frequent extranodal site of involvement for DLBCL. Its unique microbiota and persistent chronic inflammatory background can interact with and reinforce the abnormally active aerobic glycolysis (the Warburg effect) in tumor cells, forming a cancer-promoting vicious cycle[2]. This metabolic reprogramming not only fuels rapid tumor cell proliferation but also profoundly reshapes the local immune microenvironment, leading to suppressed anti-tumor immune responses[3]. Therefore, in PGI-DLBCL, glycolysis-centric metabolic reprogramming may be closely linked to the immune microenvironment, jointly influencing disease progression.

Alpha-enolase (ENO1) is a key rate-limiting enzyme in the glycolytic pathway. Its overexpression is associated with poor clinical outcomes in various malignancies[4–7]. However, its prognostic significance in the unique microenvironment of PGI-DLBCL and its relationship with intratumoral immune cell infiltration remain to be elucidated. Tumor-infiltrating lymphocytes (TILs) and tumor-associated macrophages (TAMs) are two core cellular components of the tumor immune microenvironment. Furthermore, cancer is a systemic disease. The local tumor microenvironment and the systemic peripheral immune status dynamically interact, collectively determining the ultimate anti-tumor immune outcome. The peripheral blood LMR, an easily accessible systemic immune-inflammatory marker, has demonstrated significant prognostic predictive value in various diseases[8,9].

The present work sought to explore the prognostic implication of ENO1 in PGI-DLBCL and its association with the local immune microenvironment by detecting ENO1 expression levels in tumor tissue and assessing the infiltration of TILs and TAMs. Furthermore, by concurrently analyzing the LMR, it sought to explore the correlation between ENO1 expression and this systemic immune-inflammatory status indicator.

Materials and methods

Data collection and follow-up

This study retrospectively enrolled 95 newly diagnosed and treatment-naïve PGI-DLBCL patients confirmed by the Pathology Department of the Affiliated Tumor Hospital of Xinjiang Medical University between June 2011 and January 2025. Collected data included pre-treatment peripheral blood lymphocyte and monocyte counts, age, and gender. Diagnosis was established per the WHO Classification of Haematolymphoid Tumours (5th ed.) and verified by two independent senior pathologists. The study was conducted in compliance with the Declaration of Helsinki and received ethical approval from the Xinjiang Medical University Ethics Review Committee (Approval No. K‑2025119).

LMR calculation and grouping

Peripheral blood LMR was calculated as the absolute lymphocyte count divided by the absolute monocyte count. As there is no universally accepted prognostic cut-off value for LMR, X-tile software (version 3.6.1) was used to identify an optimal overall survival‑based cut‑off of 1.5. Accordingly, patients were divided into low LMR (< 1.5) and high LMR (≥ 1.5) groups solely for correlation analyses. To avoid circular analysis bias, LMR was treated as a continuous variable in the prognostic analysis.

Immunohistochemistry and immunofluorescence staining

The following primary antibodies were used: ENO1 (Affinity Biosciences, clone DF6191), CD68 (ZSGB-BIO, clone ZM0060), CD4 (clone ZM0418), CD8 (clone ZA0508), and Foxp3 (Boster Biological Technology, clone BA2032-1). Macrophage phenotype identification was performed using multiplex immunofluorescence staining based on tyramide signal amplification (TSA) technology, employing a kit (AFIHC033) provided by Hunan Aifang Biotechnology Co., Ltd.

Interpretation criteria

Based on previous literature, three physicians independently interpreted the results. Low expression for CD4, CD8, and CD68 was defined as a comprehensive score < 4[10]. Foxp3 positivity was defined as > 20% positive cells[11]. Low ENO1 expression was defined as a comprehensive score < 8[12]. Confocal microscopy was used to capture fluorescent images of tumor areas. ImageJ 1.53a was used to count cells in the captured fluorescent images. Based on multiplex immunofluorescence detection of macrophage surface markers, cells co-expressing CD68 and CD86 were interpreted as M1-type, and cells co-expressing CD68 and CD163 were interpreted as M2-type.

Statistical analysis

Continuous variables were processed according to their distribution characteristics. Normally distributed data are presented as mean ± standard deviation, and group comparisons were made using the independent samples t-test. Non-normally distributed data are presented as median (interquartile range) [M (Q1, Q3)], and group comparisons were made using the non-parametric Mann–Whitney U test. Categorical variable comparisons were performed using the chi-square test or Fisher’s exact test. Survival analysis was conducted using the Kaplan–Meier method, and survival curves were compared using the Log-rank test. Multivariate Cox proportional hazards models were employed to identify independent prognosticators. All tests were two‑tailed, and P < 0.05 was considered statistically significant. Statistical analyses were carried out using SPSS 27.0, R 4.5.0, and GraphPad Prism 10.4.1.

Results

Clinicopathological characteristics of patients

This study included 95 newly diagnosed PGI-DLBCL patients. Their baseline characteristics are summarized in Table 1. The male-to-female ratio was nearly equal (49.5% vs. 50.5%). The median age was 64 years. Primary gastric involvement accounted for 58.95% (56/95), and the germinal center B-cell (GCB) subtype accounted for 60.00% (57/95). According to the pre-set cut-off value, 18 patients (18.95%) had low LMR (< 1.5). Lactate dehydrogenase (LDH) levels were < 240 U/L in 64.21% (61/95) of patients, and β2-microglobulin (BMG) levels were < 3.2 mg/L in 69.47% (66/95). The cell proliferation index (Ki-67) was ≤ 70% in 22.11% (21/95) of specimens. A total of 20 cases in this study were classified as double-expressor lymphoma, accounting for an overall positivity rate of 21.05% (20/95).

Expression of ENO1, TILs, and TAMs

Immunohistochemical staining of 95 PGI-DLBCL paraffin specimens showed high expression of CD4 in 58 cases (61.05%), CD8 in 66 cases (69.47%), CD68 in 62 cases (65.26%), and Foxp3 positivity in 48 cases (50.53%). Immunofluorescence results indicated the proportion of M2 cells among total macrophages was 0.37 ± 0.18, and the M2/M1 ratio was 1.5 (1.0, 3.0). Representative immunohistochemical and immunofluorescence findings are illustrated in Figs. 1 and 2.

Correlation of ENO1 with clinicopathological features, TILs, and TAMs

ENO1 was correlated with gender, LDH, and CD68⁺ macrophages (all P < 0.05, Table 2). The proportion of M2/total macrophages in the ENO1 high-expression group was 0.44 (0.37, 0.57), which was significantly higher than that in the low-expression group. Similarly, the M2/M1 ratio in the ENO1 high-expression group was 2.58 (1.53, 4.00), significantly higher than that in the low-expression group. All differences were statistically significant (P < 0.05, Table 3).There was no correlation between double‑expressor status and ENO1 expression level (P = 0.154). A possible explanation for this finding is the lack of a direct mutual regulatory relationship between ENO1 and the MYC/BCL‑2 protein pathways.

Relationship between ENO1 and the systemic immune index LMR

To investigate the interaction between metabolic reprogramming and systemic immune status, we examined the association between ENO1 expression and the peripheral immune marker LMR. LMR was lower in the high-ENO1 cohort (2.83 ± 1.68) than in the low-ENO1 cohort (3.71 ± 1.56), as shown in Fig. 3.

To further explore the association between ENO1 and the systemic immune index LMR, we analyzed the correlations between LMR and the clinicopathological features of PGI-DLBCL.LMR was associated with BMG, LDH, and CD8⁺ T-cell expression (all P < 0.05, Table 4). LMR showed no significant associations with patient gender, age, Ki-67 index, immunophenotype, CD4⁺ T-cell expression, Foxp3 expression, or CD68⁺ macrophages (all P > 0.05), as shown in Table 4.

Prognostic factor analysis

Among the 95 DLBCL patients, 33 died by the end of follow-up, with a median survival time of 86 months. The prognostic impact of gender, BMG, CD4⁺ T cells, ENO1 expression, CD8⁺ T cells, and CD68⁺ macrophages on PGI-DLBCL was evaluated using Kaplan–Meier curves and the log-rank test. Some of these factors were significantly associated with overall survival (P < 0.05), as shown in Fig. 4.

In this study, only 33 outcome events were observed. To mitigate the risk of overfitting and enhance model stability, variable selection was further performed using LASSO regression on the basis of the eight variables identified by univariate analysis. Ten-fold cross-validation was employed in the LASSO regression procedure. Taking the number of outcome events into consideration, the optimal tuning parameter lambda.1se (value of 0.189) was chosen, which resulted in the selection of three independent variables with non-zero coefficients: ENO1, CD4, and LMR. The coefficients for each variable are presented in Table 5, and the LASSO regression path selection plots are illustrated in Figs. 5 and 6.

When the selected variables were entered into a multivariate Cox regression analysis,LMR (HR = 0.543, 95% CI: 0.399–0.739, P < 0.001) and CD4⁺ T-cell infiltration (HR = 0.324, 95% CI: 0.153–0.684, P = 0.003) were identified as independent protective prognostic factors,with higher levels indicating a more favorable prognosis. Conversely, high ENO1 expression was identified as an independent unfavorable prognostic factor, associated with shorter overall survival (HR = 9.032, 95% CI: 2.170–37.595, P = 0.002) (Table 6).

Discussion

Metabolic reprogramming is a core process in tumor development. Through metabolic adaptability, tumor cells rapidly respond to oxygen and nutrient environments at different metastatic sites, competitively uptake nutrients like glucose and glutamine, and inhibit immune cell function. Additionally, the accumulation of metabolic products such as lactate and abnormal lipid metabolism can further remodel the immune microenvironment, facilitating tumor immune escape.

In this process, ENO1, a key enzyme in glycolysis and a multifunctional protein, plays a role beyond merely promoting tumor cell metabolism and proliferation. Studies have shown that in bladder cancer, high ENO1 expression can lead to CD8⁺ T-cell exhaustion[13]. In lymphohematopoietic diseases, high ENO1 expression is also associated with an immunosuppressive microenvironment, characterized by increased M2 macrophage infiltration and decreased CD8⁺ T-cell infiltration[14]. In cervical squamous cell carcinoma, ENO1-driven glycolysis intensifies the lactate-dominated acidic tumor microenvironment. This environment, by inducing pro-inflammatory cytokine secretion and upregulating HIF-1α in TAMs, drives macrophage polarization towards the M2 phenotype while attenuating T-cell activation. These findings suggest that ENO1 may be a pivotal molecule linking tumor cell metabolism to immune escape mechanisms.

The findings of this study are consistent: high ENO1 expression is associated with shortened OS, increased CD68⁺ macrophage infiltration, and a marked shift toward M2 polarization relative to low ENO1 expression. This suggests that ENO1 may have a role in recruiting macrophages and promoting their polarization towards the M2 phenotype. Although classical theory holds that M1 macrophage metabolism relies primarily on glycolysis while M2 relies more on oxidative phosphorylation[15,16], the glycolytic enzyme ENO1 in this study appeared to favor inducing M2 polarization. As previously reported, TAMs the most abundant immune cell population within TME exhibit dynamic and heterogeneous characteristics[17], with their function and polarization status regulated by metabolites present in the TME. To meet the demands of rapid proliferation, tumor cells preferentially utilize the glycolytic pathway for quick energy supply, thereby creating a TME characterized by high lactate, low glucose, and hypoxia, which in turn drives M2 polarization[18–20]. This phenomenon has been corroborated across various disease settings. When the regulatory effect of lactate on M2 macrophage polarization surpasses its role as an energy substrate, the expression of key glycolytic enzymes may become positively correlated with M2 polarization.

Effective anti-tumor immune responses depend on complex and sustained interactions between the local tumor microenvironment and the host’s systemic immune system. Locally, M2 macrophage polarization and T-cell functional exhaustion are core mechanisms driving immune escape and disease progression. However, these local immunological changes are not isolated events; their initiation and maintenance are profoundly influenced and regulated by the state of the systemic immune system.

The peripheral blood LMR is an easily measurable indicator reflecting systemic inflammatory and immune balance status. LMR has been validated as an independent prognosticator across multiple malignancies, including DLBCL, follicular lymphoma, gastric cancer, and colorectal cancer, and can predict responses to certain therapies[21–23]. This study similarly found LMR to be an prognostic factor in PGI-DLBCL, with low LMR associated with poor patient survival outcomes, consistent with previous research.

Mechanistically, this association may stem from the shaping of the local immune environment by the systemic immune status. Studies in other cancers have shown correlations between LMR and the infiltration density of specific lymphocyte subsets within tumors. For example, in oral squamous cell carcinoma, a positive correlation exists between LMR and CD4⁺ and CD20⁺ infiltrating lymphocytes[24]. Research has also demonstrated a negative correlation between LMR and CD163⁺ cells in DLBCL tissues, corroborating the link between systemic and local immunity[25]. Some propose that in DLBCL, low LMR may increase tumor-associated macrophage infiltration, leading to immunosuppression. Although no significant correlation emerged between LMR and total macrophage infiltration in the present cohort, a positive trend with intratumoral CD8⁺ T-cell levels was observed. The precise biological mechanism requires further investigation with larger sample sizes.

This study suggests that high ENO1 expression is an adverse prognostic factor in PGI-DLBCL, providing evidence for its role as an independent prognostic biomarker in this subtype. Furthermore, ENO1 expression level was correlated with peripheral blood LMR, although the underlying mechanism remains unclear. Further studies are needed to explore the biological basis of this association and to validate it in larger multicenter cohorts.

This study has several limitations. First, it is a single-center retrospective design, which may introduce bias. Second, the proportion of the GCB subtype in our cohort was higher than that of the non-GCB subtype, which is inconsistent with most previous reports on extranodal DLBCL where the non-GCB subtype is more prevalent. This discrepancy may be related to the relatively small total sample size, as a limited sample size can lead to subtype proportions that deviate from the true population distribution. Third, due to the retrospective design, clinical data such as treatment regimens and International Prognostic Index (IPI) scores were incomplete for some patients, which may affect the comprehensiveness of the prognostic analysis. Fourth, the precise molecular mechanism by which ENO1 regulates macrophage polarization has not been fully elucidated; only a correlation has been observed thus far. Fifth, the assessment of the immune microenvironment was based on only a few markers and failed to comprehensively characterize the complex immune cell network. Future studies are needed to validate and extend our findings through multicenter collaboration, larger sample sizes, improved clinical data collection, as well as basic experiments and multi-omics approaches.

Conclusions

This study suggests that ENO1 may be an independent prognostic biomarker in PGI-DLBCL, with its expression correlated with LMR and possibly involved in macrophage recruitment and polarization toward the M2 phenotype.There may be an intrinsic link among local immunity, circulating immunity, and metabolism, but the underlying mechanisms remain unclear and warrant further investigation in future studies.

References

[1] Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the world health organization classification of haematolymphoid tumours: lymphoid neoplasms. Leukemia. 2022;36(7):1720-1748.

[2] Tan C, Ma H, Chen J, et al. Critical role of IL1R2-ENO1 interaction in inhibiting glycolysis-mediated pyroptosis for protection against lethal sepsis. Adv Sci (Weinh). 2025;12(39):e02297.

[3] Chandel NS. Glycolysis. Cold Spring Harb Perspect Biol. 2021;13(5).

[4] Xu W, Yang W, Wu C, Ma X, Li H, Zheng J. Enolase 1 correlated with cancer progression and immune-infiltrating in multiple cancer types: a pan-cancer analysis. Front Oncol. 2020;10:593706.

[5] Chen S, Xu Y, Zhuo W, Zhang L. The emerging role of lactate in tumor microenvironment and its clinical relevance. Cancer Lett. 2024;590:216837.

[6] Ma Q, Jiang H, Ma L, et al. The moonlighting function of glycolytic enzyme enolase-1 promotes choline phospholipid metabolism and tumor cell proliferation. Proc Natl Acad Sci USA. 2023;120(15):e2209435120.

[7] Sun L, Suo C, Zhang T, et al. ENO1 promotes liver carcinogenesis through YAP1-dependent arachidonic acid metabolism. Nat Chem Biol. 2023;19(12):1492-1503.

[8] Xiang Q, Liu Y, Xiao J, Ou L, Du J. Prognostic value of lymphocyte-to-monocyte ratio (LMR) in patients with prostate cancer: a systematic review and meta-analysis. Am J Mens Health. 2024;18(2):15579883241234747.

[9] Tan S, Zheng Q, Zhang W, Zhou M, Xia C, Feng W. Prognostic value of inflammatory markers NLR, PLR, and LMR in gastric cancer patients treated with immune checkpoint inhibitors: a meta-analysis and systematic review. Front Immunol. 2024;15:1408700.

[10] Guo Y, Zhang X, Wu L, et al. Correlation between immune microenvironment and clinicopathological characteristics and prognosis of primary large B-cell lymphoma of immune-privileged sites. Pathol Res Pract. 2024;264:155720.

[11] de Groen RAL, de Groot FA, Böhringer S, et al. Superior survival in diffuse large B cell lymphoma of the bone with immune rich tumor microenvironment. Blood Cancer J. 2025;15(1):82.

[12] Xu X, Chen B, Zhu S, et al. Hyperglycemia promotes snail-induced epithelial-mesenchymal transition of gastric cancer via activating ENO1 expression. Cancer Cell Int. 2019;19:344.

[13] Su Z, You L, He Y, Chen J, Zhang G, Liu Z. Multi-omics reveals the role of eno1 in bladder cancer and constructs an epithelial-related prognostic model to predict prognosis and efficacy. Sci Rep. 2024;14(1):2189.

[14] Dai L, Fan G, Xie T, et al. Single-cell and spatial transcriptomics reveal a high glycolysis B cell and tumor-associated macrophages cluster correlated with poor prognosis and exhausted immune microenvironment in diffuse large B-cell lymphoma. Biomark Res. 2024;12(1):58.

[15] Van den Bossche J, Baardman J, de Winther MP. Metabolic characterization of polarized M1 and M2 bone marrow-derived macrophages using real-time extracellular flux analysis. J Vis Exp. 2015;(105):53424.

[16] Van den Bossche J, Baardman J, Otto NA, et al. Mitochondrial dysfunction prevents repolarization of inflammatory macrophages. Cell Rep. 2016;17(3):684-696.

[17] Li M, Yang Y, Xiong L, Jiang P, Wang J, Li C. Metabolism, metabolites, and macrophages in cancer. J Hematol Oncol. 2023;16(1):80.

[18] Shen C, Liu J, Jiao W, et al. A feed-forward loop based on aerobic glycolysis and TGF-β between tumor-associated macrophages and bladder cancer cells promoted malignant progression and immune escape. J Cancer Res Clin Oncol. 2023;149(14):12867-12880.

[19] Jiang H, Wei H, Wang H, et al. Zeb1-induced metabolic reprogramming of glycolysis is essential for macrophage polarization in breast cancer. Cell Death Dis. 2022;13(3):206.

[20] Liu X, Sun H, Liang J, et al. Metabolic interplay between endometrial cancer and tumor-associated macrophages: lactate-induced M2 polarization enhances tumor progression. J Transl Med. 2025;23(1):923.

[21] Xia WK, Lin QF, Shen D, Liu ZL, Su J, Mao WD. Prognostic significance of lymphocyte-to-monocyte ratio in diffuse large B-cell lymphoma: a systematic review and meta-analysis. FEBS Open Bio. 2016;6(6):558-565.

[22] Mozas P, Ould Ammar R, Chartier L, et al. A low lymphocyte-to-monocyte ratio is independently associated with early relapse (POD24) in high tumour burden follicular lymphoma: a relevance subanalysis. Br J Haematol. 2025;206(5):1380-1389.

[23] Gao F, Hu J, Zhang J, Xu Y. Prognostic value of peripheral blood lymphocyte/monocyte ratio in lymphoma. J Cancer. 2021;12(12):3407-3417.

[24] Ruiz-Ranz M, Lequerica-Fernández P, Rodríguez-Santamarta T, et al. Prognostic implications of preoperative systemic inflammatory markers in oral squamous cell carcinoma, and correlations with the local immune tumor microenvironment. Front Immunol. 2022;13:941351.

[25] Wang J, Gao K, Lei W, et al. Lymphocyte-to-monocyte ratio is associated with prognosis of diffuse large B-cell lymphoma: correlation with CD163 positive M2 type tumor-associated macrophages, not PD-1 positive tumor-infiltrating lymphocytes. Oncotarget. 2017;8(3):5414-5425.

RIGHTS & PERMISSIONS

The Author(s) 2026. This article is published by Higher Education Press at journal.hep.com.cn.