Lung cancer is the most frequently diagnosed malignancy and the leading cause of cancer death worldwide, accounting for approximately 2.5 million new cases and 1.8 million deaths in 2022 (1). Non-small cell lung cancer accounts for about 85% of cases, and adenocarcinoma is now its most common histological subtype in most regions, including East Asia (2, 3). Surgery is the primary treatment for early-stage disease, yet outcome varies considerably among patients with the same pathological stage. Although the advance in therapeutic strategies such as targeted therapy, small molecule and immune checkpoint blockade (ICB), novel therapeutic strategies are still necessary to overcome the acquired resistance. Therefore, novel therapeutic targets for drug development are needed.

α-Enolase (ENO1) is a glycolytic enzyme that catalyzes the conversion of 2-phosphoglycerate to phosphoenolpyruvate (4). In addition to this cytosolic metabolic function, ENO1 is a multifunctional protein whose role depends on its subcellular localization (5, 6). A portion of ENO1 is translocated to the plasma membrane, where it acts as a plasminogen receptor (7). Surface-bound plasminogen is converted to plasmin, which promotes pericellular proteolysis, extracellular matrix degradation, and tumor cell invasion (6, 8). This membrane translocation is regulated by post-translational modification rather than by total protein level, so the cytosolic and surface pools of ENO1 are functionally distinct (6, 8). However, most studies quantified total ENO1 and did not distinguish the surface and cytosolic ENO1, which have distinct functions. ENO1 overexpression has been reported in many cancers and is frequently associated with poor prognosis (6, 9). In lung cancer, ENO1 was identified as a tumor-associated antigen whose overexpression correlated with worse clinical outcome (10, 11), and ENO1 promoted glycolysis, growth, migration, and invasion through FAK-mediated PI3K/AKT signaling (12, 13). Notably, Hsiao et al. showed that the surface fraction of ENO1 mediates extracellular matrix degradation and metastasis in lung cancer cells, and that antibody targeting of surface ENO1 suppressed metastatic spread (8). Additionally, ENO1 has been reported to interact with hepatocyte growth factor receptor (HGFR) for HGFR/Wnt signaling activation to promote tumor growth and metastasis in lung cancer (14). Targeting surface ENO1 by chimeric anti-ENO1 mAb (chENO1-22) inhibited lung tumor metastasis and prolonged survival in animal models (14). However, whether surface and cytosolic ENO1 differ in prognostic value has not been examined in patients with LUAD. We recently demonstrated that ENO1 membrane translocation is driven by TGF-β1/Smad3 signaling through PRMT5-mediated symmetric dimethylation of ENO1 (15). Surface ENO1 then interacts with monocarboxylate transporter 4 (MCT4) to promote lactate secretion, M2 macrophage recruitment, and an immunosuppressive microenvironment, and a humanized anti-surface-ENO1 antibody (HuL001) reprogrammed macrophage polarization and improved the response to radiotherapy (11).

Therefore, we aimed to evaluate the prognostic value of surface ENO1 in LUAD patients. In this study, we evaluated surface and cytosolic ENO1 by immunohistochemistry in patients with LUAD and analyzed their relationship with clinicopathological parameters and with 5-year overall survival (OS) and progression-free survival (PFS). Here, we found that patients with high surface ENO1 were associated with lymph node metastasis and tumor differentiation. Patients with high surface ENO1 were associated with poor 5-year OS and PFS. Taken together, our studies demonstrated the prognostic value of surface ENO1 is superior to cytosolic ENO1 in 5-year PFS and OS in patients with LUAD.

Patient characteristics. A total of 174 patients with lung adenocarcinoma who underwent surgical resection at China Medical University Hospital between 2013 and 2017 were enrolled in this study. Pathological staging was based on the 7^th edition of the American Joint Committee on Cancer (AJCC) staging system, and clinical information was extracted from the electronic medical records (16). This study was reviewed and approved by the Institutional Review Board (IRB) of China Medical University Hospital [CMUH105-REC2-073]. The following clinicopathological parameters were analyzed: sex, age, smoking status, tumor side, pT status, pN status, pTNM stage, tumor differentiation, lymphovascular invasion, and EGFR mutation status (17, 18).

Immunohistochemistry. Immunohistochemical staining was performed on 3-μm formalin-fixed, paraffin-embedded tumor sections using (anti-ENO1, ab227978, Abcam, Cambridge, UK) according to the manufacturer’s instructions, and counterstained with hematoxylin (17). Surface (membranous) and cytosolic staining were scored separately by two pathologists blinded to clinical data. For each compartment, the Histo-score (H-score) was calculated as the staining intensity (0, none; 1, weak; 2, moderate; 3, strong) multiplied by the percentage of positive tumor cells, giving a range of 0 to 300 (19). The levels of surface and cytosolic ENO1 were categorized as low or high according to the median H-score (15, 20).

Statistical analysis. Statistical analysis was performed using SAS software version 9.4 (SAS Institute, Cary, NC, USA). Pearson’s chi-square and Fisher’s exact tests were used for group comparisons, with a two-sided p-value and a significance level of 0.05. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated by Cox regression analysis (21). The 5-year OS and PFS were calculated by the Kaplan–Meier method, and univariate comparisons were performed using the log-rank test. Survival endpoints were censored at 60 months (22, 23).

Patient characteristics. This study enrolled 174 resectable LUAD patients with clinicopathological characteristics presented in Table I. Overall, the mean age of these patients at diagnosis was 61.0±11.47 years (range=25-85 years), and the majority included in this study were women (58.6%). Fifty-one patients (29.3%) presented regional lymph node metastases, while seventy-four patients (42.5%) had lymphovascular invasion (Table I).

Surface ENO1 expression is associated with adverse clinicopathological features. The relationship between ENO1 expression and clinicopathological parameters is presented in Table I. The expression of ENO1 was evaluated in Figure 1A. High surface ENO1 expression was significantly associated with older age (p=0.044), nodal metastasis (p=0.030, Figure 1B), and poor tumor differentiation (p=0.017, Figure 1C). Sex, smoking status, tumor side, pT status, pTNM stage, lymphovascular invasion, and EGFR mutation status were not associated with surface ENO1. In contrast, cytosolic ENO1 expression was not associated with any clinicopathological parameter. Surface and cytosolic ENO1 were positively correlated with each other (p<0.001).

High surface ENO1 expression is associated with poor 5-year overall survival (OS) and progression-free survival (PFS). Kaplan–Meier analysis showed that patients with high surface ENO1 had significantly worse 5-year OS (64.0% vs. 80.1%, log-rank p=0.041) and 5-year PFS (44.8% vs. 65.0%, log-rank p=0.047) than those with low expression (Table II, Figure 2A-B). In contrast, cytosolic ENO1 was not associated with 5-year OS (67.1% vs. 77.4%, p=0.337) or PFS (49.0% vs. 61.0%, p=0.343, Table II).

Surface ENO1 was not an independent prognostic factor for LUAD patients. In univariate Cox regression, high surface ENO1 expression was associated with an increased risk of 5-year OS (HR=1.94, 95%CI=1.02-3.71, p=0.045) and of 5-year PFS (HR=1.62, 95%CI=1.00-2.63, p=0.049) (Table III). Nodal metastasis, advanced pTNM stage, and poor differentiation were the strongest prognostic factors, whereas cytosolic ENO1 and EGFR status were not associated with survival. These results indicate that tumor surface ENO1 expression has a significant prognostic value for LUAD patients. Subsequently, we examined whether the inclusion of other variables affected the parameter estimate for tumor surface ENO1 (Table IV). We found significant difference in surface ENO1 status after adjusted with clinical parameters.

In this study, we found that surface ENO1, but not cytosolic ENO1, was associated with worse pathological features, poor 5-year OS and PFS in patients with LUAD. By stratifying subcellular localization, we found that surface ENO1 was associated with powerful prognostic value, whereas the cytosolic ENO1 was not informative. These results demonstrated that surface ENO1 was not an independent prognostic factor for LUAD patients.

The association of surface ENO1 with nodal metastasis and poor differentiation is consistent with previous studies (4, 7, 9). As a plasminogen receptor, surface ENO1 promotes plasmin generation, extracellular matrix degradation, and invasion in colorectal and lung cancer (14, 15). In addition to its role in plasmin generation, our previous work further showed that surface ENO1 interacts with MCT4 to drive lactate secretion and M2 macrophage recruitment, establishing an immunosuppressive microenvironment that reduces the response to radiotherapy (15). Moreover, ENO1 promotes cancer metastasis via stimulating metabolic reprogramming in osteosarcoma (24). These results suggested that ENO1 may regulate cancer cell metabolism and create an immunosuppressive tumor microenvironment to promote cancer metastasis. Additionally, ENO1 also promotes cancer metastasis via diverse mechanisms such as HGFR/WNT signaling–driven epithelial-to-mesenchymal transition (14). Li et al. indicated targeting surface ENO1 by chimeric anti-ENO1 antibodies significantly inhibit tumor growth and cancer metastasis (14). These results indicated that surface ENO1 participated in cancer progression and metastasis, supporting that surface ENO1 was a prognostic factor in LUAD patients. Moreover, the finding that cytosolic ENO1 was associated with neither clinicopathological features nor survival supports the interpretation that the prognostic effect is specific to the surface-localized ENO1 protein.

Previous studies also demonstrated that total ENO1 might be a marker for different malignancies, such as glioma (25-27) and lung cancer (13). However, the prognostic value of ENO1 in lung cancer was inconsistent with total ENO1 reported as both down-regulated (28) and overexpressed (13) in NSCLC. Our results suggest that this discrepancy may be explained by the different compartment. If the surface ENO1 is the biologically relevant fraction, whole-cell quantification of ENO1 would be expected to give inconsistent associations. Therefore, distinguishing surface and cytosolic ENO1 may provide a more reliable assessment of prognosis in different malignancies. Moreover, surface ENO1 is more accessible by antibody-based targeting, as shown for HuL001 and chENO1-22 in preclinical models (14, 15). Cappello et al. developed an anti-ENO1 DNA vaccine that can deliver multiple anti-tumor immune effects (29, 30). Anti-ENO1 DNA treatment combined with gemcitabine (GEM) unleashed CD4⁺ anti-tumor activity and strongly impaired tumor progression, suggesting that targeting surface ENO1 is capable of overcoming tolerance and generating a significant anti-tumor immune response. Therefore, identifying the prognostic value of surface ENO1 protein was critical. But this study has several limitations. This retrospective study was conducted at a single center, and the survival associations require external validation. In the multivariate model, the surface-ENO1 hazard ratio was attenuated after adjustment for tumor differentiation and stage. Because these features may lie downstream of surface-ENO1-driven invasion rather than act as independent confounders, we interpret this attenuation with caution and do not claim that surface ENO1 is an independent prognostic factor from the present data. This relationship cannot be established from cross-sectional material and requires prospective and mechanistic confirmation. Finally, the reproducibility of compartment-specific H-scoring across laboratories should be established before clinical application.

In conclusion, our results showed that high surface ENO1 expression is associated with poor differentiation, nodal metastasis, and poor 5-year survival in patients with LUAD. These findings suggest that surface ENO1 could serve as a prognostic biomarker and a therapeutic target in LUAD.

The Authors claim that there are no conflicts of interest.

Data curation: Yuan-Yao Tsai, Tao-Wei Ke, Chi-Hsien Huang; Formal analysis: Yuan-Yao Tsai, Chia-Yu Tsai, Yu-Sen Lin and Te-Hong Chen; Funding acquisition: K. S. Clifford Chao and Kevin Chih-Yang Huang; Investigation: William Tzu-Liang Chen, K. S. Clifford Chao, Kevin Chih-Yang Huang; Methodology: Kevin Chih-Yang Huang; Project administration: Kevin Chih-Yang Huang; Resources: William Tzu- Liang Chen, Tao-Wei Ke and K. S. Clifford Chao; Supervision: Kevin Chih-Yang Huang; Writing; Kevin Chih-Yang Huang.

The Authors are grateful for the tissue microarray support from the Translation Research Core, China Medical University Hospital. This study was supported in part by China Medical University Hospital (DMR-115-106). This study was partially based on clinical information from the China Medical University Hospital Cancer Registry. Experiments and data analysis were performed in part through the use of the Medical Research Core Facilities Center, Office of Research & Development at China Medical University, Taichung, Taiwan, R.O.C.

During the preparation of this manuscript, a large language model (Claude) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.