Colorectal cancer (CRC) ranks as the second most frequently diagnosed malignancy in adult women and the third in adult men, standing as the fourth leading cause of cancer-related mortality and accounting for approximately 9.2% of all cancer deaths globally (Bray et al., 2018; Dekker et al., 2019). Fewer than half of CRC cases are diagnosed at a locally advanced stage. The disease is highly heterogeneous and involves numerous genetic or somatic mutations (Zygulska and Pierzchalski, 2022). Its onset, progression, and prognosis are closely associated with the immune status of the tumor microenvironment (TME) and the metabolic activity of tumor cells (Tomasini et al., 2021; Ferkel et al., 2025; Liu et al., 2025). In recent years, increasing attention has been directed toward the roles of immune checkpoint molecules and DNA damage repair proteins in tumor progression.

Clinico-pathological characteristics such as tumor stage are well-established prognostic indicators for predicting patient survival (Brenner et al., 2014; Renfro et al., 2014). Leukocyte Immunoglobulin-Like Receptor B2 (LILRB2) is a transmembrane glycoprotein structurally similar to the immune checkpoint protein programmed cell death protein-1 (PD-1) and its ligand PD-L1. It is widely expressed on dendritic cells, macrophages, and other myeloid cells, where it inhibits immune cell activity and participates in important pathological processes, including modulation of the TME, immune evasion, and tumor progression (Chen et al., 2018). According to previous research, LILRB2 contributes to immune suppression and tumor advancement in malignancies such as acute myeloid leukemia and non-small cell lung cancer (Zhao et al., 2024). Moreover, LILRB2 expression is markedly upregulated in CRC, where its elevation correlates with unfavorable clinicopathological parameters, including poor-to-moderate differentiation, lymph node metastasis (LNM), advanced TNM stage, and poor prognosis (Wang et al., 2023). Analyses of T-cell subsets in CRC and lung adenocarcinoma have further revealed that LILRB2 overexpression is associated with decreased CD3+ and CD8+ T-cell levels and increased FOXP3+ regulatory T-cell infiltration within the TME (Cao et al., 2025).

Apurinic/apyrimidinic endonuclease 1 (APE1), a key enzyme in the base excision repair pathway, also functions as a redox-sensitive modulator that activates multiple transcription factors (Oliveira et al., 2022; Xue and Demple, 2022; Zhang et al., 2022). APE1 has been implicated in CRC progression, influencing tumor growth, chemotherapeutic resistance, and metastatic potential in both human CRC cells and carcinogen-induced CRC models. Elevated APE1 expression is closely linked to advanced disease stage and reduced survival, suggesting its potential as a prognostic and predictive biomarker for CRC (Hong et al., 2023).

Although LILRB2 and APE1 have been independently identified as regulators of tumor immunity and DNA repair, respectively, potential crosstalk or synergistic regulation between them in CRC remains unclear. Considering their associations with tumor pathology and patient prognosis, both molecules represent promising biomarkers. This study, therefore, integrates LILRB2 and APE1 within a unified analytical framework to investigate their co-expression patterns in CRC, their relationships with clinicopathological characteristics, and their combined predictive value for patient prognosis.

CRC remains one of the most prevalent malignancies of the human digestive tract. Its development follows a multistep process, beginning with benign adenomatous polyps of the large intestine and rectum that gradually transform into invasive carcinoma over time (Wu et al., 2020). Conventional treatments for CRC primarily include surgery, chemotherapy, and radiotherapy, often used in combination depending on individual patient needs. In recent years, immunotherapy has emerged as a promising therapeutic strategy for CRC (Johdi and Sukor, 2020). In this study, serum expression levels of LILRB2 and APE1 were analyzed in CRC patients to explore their correlations with clinicopathological features and prognosis. The results demonstrated that both LILRB2 and APE1 were significantly upregulated in CRC patients compared with healthy controls, suggesting that these biomarkers may play important roles in CRC onset and progression.

Specifically, LILRB2 expression was significantly associated with patient age, TNM stage, and LNM. LILRB2 is known to inhibit immune cell activity and modulate the tumor immune microenvironment through its interaction with ligands such as HLA-G, angiopoietin-like protein 2, and semaphorin-4A, thereby promoting tumor cell proliferation and metastasis (Cai et al., 2019). Consistent with earlier findings (Chen QY. et al., 2021), the current findings revealed that elevated LILRB2 levels were positively correlated with immune checkpoint proteins and LNM, contributing to tumor immune evasion and progression. Consistent with earlier evidence, overexpression of LILRB2—particularly in concert with its ligand angiopoietin-like protein 2—has been identified as a strong predictor of poor prognosis in CRC (Zhao et al., 2024). Additionally, the present results suggest that LILRB2 is relevant across different age groups, supporting its potential as a prognostic biomarker for CRC progression.

Similarly, APE1 expression was closely associated with tumor diameter, TNM stage, and LNM. Previous studies have demonstrated a strong relationship between APE1 and the tumor immune microenvironment (Li et al., 2021). APE1 expression is markedly elevated in CRC and metastatic lymph nodes compared with normal colorectal mucosa and non-metastatic lymph nodes. Its overexpression correlates with adverse pathological parameters, including lymphovascular and perineural invasion, deeper tumor infiltration, distant metastasis, and reduced overall survival (Hong et al., 2023). Consistent with these observations, our study found that patients with larger tumor diameters, advanced TNM stages, and LNM exhibited higher APE1 expression levels, further underscoring its role in CRC progression and metastasis.

Moreover, survival analysis and Cox regression models demonstrated that high expression levels of LILRB2 and APE1, advanced TNM stage, and LNM were independent risk factors for poor prognosis in CRC patients. These findings highlight the importance of incorporating these biomarkers into CRC risk stratification and therapeutic decision-making. Patients with elevated LILRB2 and APE1 expression exhibited significantly lower survival rates than those with low expression, indicating their potential as prognostic indicators for CRC outcomes. ROC analysis revealed that the combined predictive model incorporating both LILRB2 and APE1 achieved a higher AUC than either marker alone, suggesting superior prognostic accuracy when both biomarkers are evaluated in tandem. Collectively, these results suggest that LILRB2 and APE1 may serve as complementary biomarkers for assessing disease progression and predicting prognosis in CRC patients.

It is well recognized that the TNM staging system plays an irreplaceable role in determining CRC prognosis (Chen K. et al., 2021). However, as an anatomical localization-based system, it cannot fully capture the biological heterogeneity inherent to tumors (Jiang et al., 2025). In the present study, it was observed that even among patients with identical TNM stages, significant variations exist in the expression levels of LILRB2 and APE1, which could further distinguish subgroups with different survival outcomes. This finding suggests that the molecular information conveyed by LILRB2 and APE1 extends beyond the morphological boundaries defined by the TNM classification. Therefore, integrating these two molecular markers with TNM staging may enable the development of a more dynamic and predictive “anatomical-molecular” composite model for CRC prognosis. Moreover, microsatellite instability (MSI) is an important molecular phenotype in CRC, closely associated with both prognosis and responsiveness to immunotherapy (Greco et al., 2023; Jonchere et al., 2024). The present study provides new insights into the potential mechanistic links between LILRB2 and APE1 expression and MSI status in CRC. Specifically, LILRB2 may characterize a myeloid immunosuppression–dominant tumor microenvironment within microsatellite-stable (MSS) subtypes, offering a potential therapeutic target for overcoming the limited efficacy of immunotherapy in this group. Conversely, the DNA repair function of APE1 in MSI-high (MSI-H) tumors may undergo adaptive modulation, reflecting its role in maintaining genomic stability under immune pressure. Exploring the interplay between these two molecules and MSI status could significantly enhance the precision of immune-based molecular typing and the optimization of personalized treatment strategies for CRC.

Despite its novel findings, this study is not without limitations. First, it is a single-center investigation with a relatively small sample size and a follow-up period of only two years. These factors may limit the generalizability of the results and restrict the evaluation of long-term prognostic outcomes, such as the 5-year survival rate. Second, the current analysis focused solely on serum samples; validation of LILRB2 and APE1 protein expression at the tissue level through immunohistochemistry (IHC) was not performed. The lack of direct tissue-level verification somewhat weakens the evidence linking these biomarkers to the local tumor microenvironment. Most importantly, the study primarily elucidates the clinical correlations and predictive value of LILRB2 and APE1, without exploring the underlying molecular mechanisms through which they promote CRC progression.

In conclusion, the study confirms that serum levels of LILRB2 and APE1 are markedly elevated in CRC patients and are closely associated with key pathological characteristics and prognosis. These molecules represent promising biomarkers for CRC risk stratification, early diagnosis, and therapeutic planning. The novelty of this research lies in the conceptual and empirical integration of two functional molecules—originating from distinct biological pathways—into a synergistic predictive framework. This combined model demonstrated superior prognostic performance, with a higher AUC than either single indicator, thereby providing strong evidence for the refinement of CRC prognostic evaluation systems. To overcome current limitations, future studies should involve large-scale, multicenter, prospective cohorts with extended follow-up durations to validate the clinical utility of combined LILRB2 and APE1 detection. Furthermore, IHC analyses of archived tumor tissues could be conducted to verify expression localization and clinical correlations at the tissue level. Complementary in vitro experiments should explore the regulatory interactions and mechanistic roles of LILRB2 and APE1 in modulating the malignant phenotype of CRC cells. Ultimately, the establishment of a comprehensive prognostic prediction system that integrates clinical staging, molecular biomarkers, and molecular subtyping may enable individualized risk stratification and guide precision therapy in CRC.