cPLA2α inhibition potentiates anti-PD-1 immunotherapy in lung adenocarcinoma via remodeling tumor microenvironment Authors: Ran Zuo, Yuchao He, Yulin Wu, Huien Yang, Yimeng Wang, Yu Wang, Wenchen Gong, Dongming Liu, Gengwei Huo, Liwei Chen, Yi Luo, Aomei Ling, Zhanyu Pan, Zhiyong Liu, Peng Chen, Hua Guo Published: 2026 ABSTRACT Although immune checkpoint inhibitors (ICIs) against PD-1/PD-L1 have revolutionized lung adenocarcinoma (LUAD) therapy, resistance and low response rates persist. Cytoplasmic phospholipase A2α (cPLA2α), a key enzyme in inflammatory pathways, promotes tumor progression, yet its immunomodulatory role in LUAD is unclear. Here, cPLA2α overexpression is found in LUAD tissues, correlating with poor prognosis. cPLA2α knockdown inhibits lung cancer cell proliferation, migration in vitro, and tumor growth in immunodeficient mice, but efficacy diminishes in immunocompetent models due to STAT3-mediated PD-L1 upregulation. In syngeneic models, cPLA2α silencing promotes CD8+ T cell and M1 macrophage infiltration and reduces immunosuppressive neutrophils and M2 macrophages. Strikingly, cPLA2α knockdown with anti-PD-1 synergistically suppresses tumor growth, where granzyme B+ cytotoxic CD8+ T cells amplifies. Our findings unveils cPLA2α’s dual role in LUAD: promoting tumorigenesis and orchestrating immune evasion via STAT3–PD-L1 signaling. Targeting cPLA2α may overcome PD-1 blockade resistance, particularly benefiting patients with high cPLA2α and low baseline PD-L1. This study highlights cPLA2α as a promising therapeutic target to enhance ICI efficacy and reshape LUAD immunotherapy strategies. Introduction Lung adenocarcinoma (LUAD), the most prevalent pathological subtype of non-small cell lung cancer (NSCLC), remains a leading cause of cancer-related mortality worldwide1. Immune checkpoint inhibitors (ICIs), such as programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1), have shown remarkable potential in LUAD treatment, offering durable clinical responses and survival benefits to patients2. The successful outcomes of multiple clinical trials have expanded the application of ICIs from later-line to first-line therapies and from advanced to perioperative settings3–6. Despite these significant advancements, the clinical implementation of ICIs in LUAD treatment still encounters numerous challenges. First, the response rate to immunotherapy remains low, with only approximately 20–30% of patients benefiting from monotherapy. Second, the emergence of primary and acquired resistance severely restricts the long-term efficacy of ICIs7,8. Immune infiltration within the tumor microenvironment is regulated by a variety of complex factors, including the intrinsic immunogenicity of tumor cells, tumor angiogenesis, presence of immunosuppressive immune cells, and poor infiltration and impaired function of effector immune cells9,10. Considering these characteristics, current research on immunotherapy for LUAD focuses on two key strategies: one is to modulate the tumor microenvironment to enhance the infiltration and activation of immune cells, and the other is to identify reliable biomarkers that can accurately predict the response to immunotherapy. These strategies serve as the foundation for developing novel combination therapies based on ICIs and hold the key to overcoming immune resistance and improving treatment outcomes. Abnormal PD-L1 expression in tumor cells is a major contributor to tumor immune escape because it inhibits T cell activation and impairs anti-tumor immunity, ultimately leading to reduced immunotherapy efficacy11. Blocking the PD-1/PD-L1 interaction can restore the cytotoxic function of T cells. Increasing studies have confirmed that PD-1/PD-L1 inhibitors combined with chemotherapy, radiotherapy, angiogenesis inhibitors, and other ICIs exert superior anti-tumor efficacy and high response rate12. Our previous research also demonstrated that in epidermal growth factor receptor (EGFR)-mutant LUAD cells, PD-L1 promotes autophagy, contributes to primary resistance to EGFR-tyrosine kinase inhibitors (TKIs), and enhances anti-tumor efficacy when used in combination therapy13. This finding highlights the crucial role of PD-L1 in intracellular signal transduction independent of immunomodulatory effects and emphasizes its value as a predictive and prognostic biomarker in specific populations. It also underscores the urgent need to develop novel combination strategies to overcome immunotherapy resistance. Cytoplasmic phospholipase cPLA2α (PLA2G4A) is the rate-limiting enzyme in the production of eicosanoid compounds and plays a key role in inflammatory responses and cancer development through the arachidonic acid cascade14. Previous studies have demonstrated that cPLA2α promoted malignancy in hepatocellular and breast cancers by mediating epithelial-mesenchymal transition, and drived metastasis in cervical cancer by regulating cancer stemness15–17. In NSCLC, the upregulation of cPLA2α is associated with drug resistance and poor prognosis18. These studies predominantly focused on the direct effects of cPLA2α on tumor cells, while its effect on the host immune system and tumor microenvironment remains poorly understood. Notably, recent studies have indicated that NSCLC patients with high cPLA2α expression showed poorer responses to neoadjuvant therapy combining anlotinib and PD‑1 inhibitors, accompanied by increased infiltration of regulatory T cells (Tregs)19. Although the role of cPLA2α in the tumor microenvironment has gradually attracted attention20–22, studies on its function in tumor immunity, particularly in tumor immunotherapy, are rare and sporadic. This study systematically investigated the effect of abnormal cPLA2α expression on the tumor microenvironment of LUAD. We also explored the in vivo anti-tumor effect of cPLA2α downregulation in combination with PD-1 blockade. Our findings suggest a therapeutic strategy and theoretical basis for enhancing immunotherapy efficacy by targeting cPLA2α in patients with LUAD. Results cPLA2α is upregulated in LUAD tissues and correlated with poor prognosis Our previous studies showed that cPLA2α was highly expressed in hepatocellular carcinoma, breast cancer, and cervical cancer tissues and closely related to poor prognosis15–17. In the current work, 182 LUAD samples were evaluated by tissue microarray and immunohistochemistry (IHC) staining to further investigate the expression pattern of cPLA2α and the relationship between its expression levels and the prognosis of patients with LUAD. Representative photomicrographs of different cPLA2α expression levels are shown in Fig. 1A. More than half of the LUAD samples exhibited moderate or high cPLA2α staining (Fig. 1B). We then explored the relationship between cPLA2α protein expression and patient survival. The patients were divided into high (n = 113) and low (n = 69) cPLA2α expression groups according to their cPLA2α protein levels. Kaplan–Meier survival curves demonstrated that the patients with high cPLA2α expression exhibited poor disease-free survival (DFS) (P = 0.005, Fig. 1C) and overall survival (OS) (P < 0.001, Fig. 1D). Those with low cPLA2α expression had a higher 5-year OS rate (84.4%) than those with high cPLA2α protein expression (54.3%). Cox regression analysis indicated that increased cPLA2α expression was an independent risk factor for the OS of patients with LUAD (Supplementary Data 1). We further analyzed the correlations between various clinicopathological factors and patient prognosis. Kaplan–Meier survival curves suggested that patients with a history of smoking, large tumor size (size >3 cm), lymph node (LN) metastasis, and TNM stage II–III had poor DFS and OS (Supplementary Fig. 1A–H). Advanced age (>60), male gender, and elevated carcinoembryonic antigen (CEA) level (>5 μg/L) were also associated with short OS, but their impact on DFS was limited (Supplementary Fig. 1I–N). Forest plots of the multivariate analysis results indicated that cPLA2α could serve as a specific biomarker for patients with LUAD (Supplementary Fig. 1O). Considering the important role of cPLA2α in LUAD progression, we further analyzed the correlations between various clinical indicators and patient prognosis and found that cPLA2α expression was markedly correlated with smoking history (P = 0.001), CEA (P = 0.022), tumor size (P = 0.014), and TMN stage (P = 0.049) (Supplementary Data 2). The percentage of patients with high cPLA2α expression was significantly higher among the high-risk subgroups, including those with a smoking history (P = 0.001), elevated CEA (>5 µg/L) (P = 0.022), large tumor size (>3 cm) (P = 0.014), or advanced TNM stage (II and III) (P = 0.049), compared with their corresponding low-risk subgroups (Fig. 1E, G, I, K).Fig. 1cPLA2α expression is elevated in LUAD tissue samples and predicts poor survival in LUAD patients.A Representative images of cPLA2α expression in 182 LUAD patient samples stained with IHC (Scale bar = 100 μm and 200 μm). B The proportions of LUAD patients with different cPLA2α expression levels. C, D Kaplan–Meier survival curve showing the DFS and OS between cPLA2α low and high expression groups. E Correlation analysis between the cPLA2α IHC score and smoking status. F Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in smoker subgroup. G Correlation analysis between the cPLA2α IHC score and CEA level. H Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in CEA > 5 μg/L subgroup. I Correlation analysis between the cPLA2α IHC score and tumor size. J Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in tumor size >3 cm subgroup. K Correlation analysis between the cPLA2α IHC score and TNM stage. L Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in TNM stage II–III subgroup. M Correlation analysis between the cPLA2α IHC score and LN metastasis. N Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in LN metastasis subgroup. (O Western blotting showing the differences in the cPLA2α protein levels between eight pairs of LUAD tissue and adjacent normal tissue (n = 8). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, ****P < 0.0001. n indicates biologically independent samples. cPLA2α expression is elevated in LUAD tissue samples and predicts poor survival in LUAD patients. A Representative images of cPLA2α expression in 182 LUAD patient samples stained with IHC (Scale bar = 100 μm and 200 μm). B The proportions of LUAD patients with different cPLA2α expression levels. C, D Kaplan–Meier survival curve showing the DFS and OS between cPLA2α low and high expression groups. E Correlation analysis between the cPLA2α IHC score and smoking status. F Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in smoker subgroup. G Correlation analysis between the cPLA2α IHC score and CEA level. H Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in CEA > 5 μg/L subgroup. I Correlation analysis between the cPLA2α IHC score and tumor size. J Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in tumor size >3 cm subgroup. K Correlation analysis between the cPLA2α IHC score and TNM stage. L Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in TNM stage II–III subgroup. M Correlation analysis between the cPLA2α IHC score and LN metastasis. N Kaplan–Meier survival curve showing the OS between cPLA2α high and low expression groups in LN metastasis subgroup. (O Western blotting showing the differences in the cPLA2α protein levels between eight pairs of LUAD tissue and adjacent normal tissue (n = 8). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, ****P < 0.0001. n indicates biologically independent samples. We further probed the differential prognosis between patients with high or low cPLA2α expression within these high-risk subgroups. In all four subgroups, the patients with low cPLA2α expression exhibited better prognosis than those with high expression (Fig. 1F, H, J, L). Although there was no significant correlation between cPLA2 expression and LN metastasis (P = 0.099) (Fig. 1M), in the LN metastasis subgroup, patients with low expression of cPLA2α had a higher OS (P = 0.038) (Fig. 1N). However, no significant correlation was found between cPLA2α expression and age and sex (Supplementary Data 2). Consistently, Western blotting analysis of the eight paired LUAD tumor and adjacent nonmalignant tissue samples indicated that the LUAD tissue samples had higher cPLA2α expression than the adjacent normal tissue samples (Fig. 1O). To verify these findings, we analyzed publicly available datasets from online databases (TCGA and CTPAC). The results revealed that cPLA2α expression was higher in tumor tissues than in their corresponding normal tissues (Supplementary Fig. 1P, Q). In addition, high cPLA2α expression was associated with poor OS among patients with LUAD as revealed using the TCGA dataset (Supplementary Fig. 1R). In summary, these results suggested that cPLA2α was upregulated in LUAD tissues and might be a contributing factor and a potential therapeutic target in LUAD. cPLA2α promotes the proliferation and motility of LUAD cells, and its knockdown suppresses LUAD tumorigenesis in an immune-deficiency mouse model To further investigate the role of cPLA2α in the tumor progression of LUAD, we selected two human NSCLC cell lines (H1299 and A549 cells) to construct cell lines with stable cPLA2α upregulation (cPLA2α) and cPLA2α downregulation (sh-cPLA2α). Corresponding empty vector controls (vector and sh-Ctrl) were established for these two cell lines. Stable cPLA2α knockdown and control cells were also obtained using the mouse lung cancer cell line (LLC cell). The efficiencies of cPLA2α overexpression and knockdown were confirmed by Western blotting and qRT-PCR (Fig. 2A, B). Cell counting kit-8‌ (CCK-8) assays showed that the proliferation rate increased in the H1299 cells overexpressing cPLA2α but decreased in the A549 cells with cPLA2α knockdown compared with that in the control groups (Fig. 2C, D). The effect of cPLA2α on cell viability was also verified in mouse LLC sh-cPLA2α cells (Fig. 2E). Thus, cPLA2α was significantly associated with the proliferation ability of LUAD cells. Transwell assays further demonstrated that cPLA2α upregulation enhanced the invasion capacity of H1299 cells, and its downregulation inhibited the invasion capacity of A549 cells (Fig. 2F, G). Similar results were observed in LLC cells (Fig. 2H), suggesting that cPLA2α might promote the migration capacity of LUAD cells. These results suggested that cPLA2α might promote the malignant biological behaviors of LUAD cells.Fig. 2cPLA2α promotes proliferation and motility of LUAD cells in vitro, downregulation of cPLA2α inhibits tumor growth in immune-deficiency mouse model.Construction of H1299 cPLA2α-upregulation (cPLA2α), A549 cPLA2α-downregulation (sh-cPLA2α), and LLC cPLA2α-downregulation (sh-cPLA2α) cell lines. Western blotting A and RT-PCR B were performed to verify the expression of cPLA2α (n = 3). C–E CCK-8 assay showing the impact of cPLA2α overexpression or knockdown on H1299, A549 and LLC cell growth (n = 3–5). F–H Chemotaxis assays detected the impact of abnormal cPLA2α expression on H1299, A549 and LLC cell motility (n = 3) (Scale bar = 100 μm). I Representative images of tumors from BALB/c Nude mice in LLC sh-Ctrl and sh-cPLA2α groups. J, K Tumor volumes and weights in sh-Ctrl and sh-cPLA2α groups (n = 5). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. cPLA2α promotes proliferation and motility of LUAD cells in vitro, downregulation of cPLA2α inhibits tumor growth in immune-deficiency mouse model. Construction of H1299 cPLA2α-upregulation (cPLA2α), A549 cPLA2α-downregulation (sh-cPLA2α), and LLC cPLA2α-downregulation (sh-cPLA2α) cell lines. Western blotting A and RT-PCR B were performed to verify the expression of cPLA2α (n = 3). C–E CCK-8 assay showing the impact of cPLA2α overexpression or knockdown on H1299, A549 and LLC cell growth (n = 3–5). F–H Chemotaxis assays detected the impact of abnormal cPLA2α expression on H1299, A549 and LLC cell motility (n = 3) (Scale bar = 100 μm). I Representative images of tumors from BALB/c Nude mice in LLC sh-Ctrl and sh-cPLA2α groups. J, K Tumor volumes and weights in sh-Ctrl and sh-cPLA2α groups (n = 5). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. Considering that cPLA2α promoted the proliferation and metastasis of LUAD cells in vitro, we hypothesized that knocking down cPLA2α might also suppress the tumorigenesis of LUAD cells in vivo. To testify this hypothesis, we implanted LLC cells with stable cPLA2α downregulation (sh-cPLA2α) or expressing a control vector (sh-Ctrl) into immunodeficient BALB/c Nude mice. Tumor volume was measured every other day from day 12 post-injection. All the mice were sacrificed at the end of the 4th week, and the primary tumors were shown in Fig. 2I. Tumor volumes and weights were significantly reduced in the sh-cPLA2α group compared with those of the control group (Fig. 2J, K). Overall, these results suggested that downregulating cPLA2α suppressed LUAD tumor growth in an immune-deficient mouse model. cPLA2α downregulation leads to a compromised anti-tumor effect in immune-competent mice To further investigate the role of cPLA2α in the immune microenvironment of LUAD, we analyzed the correlation between the expression of cPLA2α and the abundance of tumor-infiltrating immune cells using the TISIDB online analysis tool. LUAD exhibit some differences from most cancers. In particular, cPLA2α expression is negatively correlated with the infiltration levels of most immune cells, including CD4+ T cells, CD8+ T cells, Tregs, natural killer (NK) cells, dendritic cells (DCs), macrophages, and myeloid-derived suppressor cells (MDSCs) (Supplementary Fig. 2A). To verify these results, we extracted the LUAD dataset from the TCGA database and analyzed it using R Studio. Among the 24 immune cell types, cPLA2α expression was significantly negatively correlated with the infiltration of T cells, Treg cells, NK cells, DCs, and macrophages (Supplementary Fig. 2B). We further investigated the association of cPLA2α expression with immunostimulators and major histocompatibility complex (MHC) molecules across various cancer types, including LUAD. As depicted in Supplementary Fig. 2C–F, cPLA2α expression was negatively correlated with most immunostimulatory molecules (such as CD40, CD70, CD86, and ICOS) and MHC molecules. These findings further suggested that high cPLA2α expression was associated with adverse immune infiltration in LUAD. CXCL9, -10, and -11/CXCR3 are known to recruit cytotoxic T lymphocytes (CTLs), NK cells, and macrophages to tumor sites23,24. Finally, we analyzed the correlation between the expression of cPLA2α and the mRNA levels of chemokines and their receptors. As shown in Supplementary Fig. 2G, cPLA2α expression was inversely correlated with CXCL9, CXCL10, CXCL11, and their receptor CXCR3, indicating that cPLA2α might play a negative regulatory role in T cell immunity and inflammatory responses. Overall, these results suggested that cPLA2α played a negative regulatory role in anti-tumor immunity in LUAD. To assess whether an intact immune system could play a role in the anti-tumor effect mediated by cPLA2α downregulation, we subcutaneously inoculated LLC cells with stable cPLA2α downregulation or control cells into immune-competent C57BL/6 mice. According to previous database analysis (Supplementary Fig. 2) and the results from immune-deficient mice (Fig. 2I–K), one would expect that the anti-tumor effect of cPLA2α downregulation would be highly pronounced in immune-competent mice. Surprisingly, no significant differences in tumor growth and weights were observed between the sh-cPLA2α and sh-Ctrl groups (Fig. 3A–C).Fig. 3Complete immune system impairs the anti-tumor efficacy of cPLA2α Knockdown in vivo.A Representative images of tumors from C57BL/6 mice in LLC sh-Ctrl and sh-cPLA2α groups. B, C Tumor volumes and weights in sh-Ctrl and sh-cPLA2α groups (n = 5). Flow cytometric analysis of tumor-infiltrating lymphocyte from C57BL/6 mice in sh-Ctrl and sh-cPLA2α groups (n = 3–5). Representative flow analysis and quantification of CD45+ immune cells (n = 5) (D), CD3+ T cells (n = 5) (E), CD8+ T cells, CD4+ T cells (n = 5) (F), GZMB+CD8+ T cells (n = 3–4) (G), Treg cells (n = 3) (H) and NK cells (n = 4) (I). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001. n indicates biologically independent samples. Complete immune system impairs the anti-tumor efficacy of cPLA2α Knockdown in vivo. A Representative images of tumors from C57BL/6 mice in LLC sh-Ctrl and sh-cPLA2α groups. B, C Tumor volumes and weights in sh-Ctrl and sh-cPLA2α groups (n = 5). Flow cytometric analysis of tumor-infiltrating lymphocyte from C57BL/6 mice in sh-Ctrl and sh-cPLA2α groups (n = 3–5). Representative flow analysis and quantification of CD45+ immune cells (n = 5) (D), CD3+ T cells (n = 5) (E), CD8+ T cells, CD4+ T cells (n = 5) (F), GZMB+CD8+ T cells (n = 3–4) (G), Treg cells (n = 3) (H) and NK cells (n = 4) (I). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001. n indicates biologically independent samples. To investigate the impact of cPLA2α downregulation in LUAD cells on the tumor immune response, we used flow cytometry to analyze the infiltration of immune cells in tumors obtained from mice injected with either sh-cPLA2α or sh-Ctrl LUAD cells at day 24. The main gating strategies for flow cytometry data analysis were shown in Supplementary Fig. 3. No significant difference was observed in the total number of tumor-infiltrating CD45+ immune cells (Fig. 3D). Meanwhile, the proportions and densities of tumor-infiltrating CD3+ T cells and CD8+ T cells increased, and the proportion of CD4+ T cells decreased slightly (Fig. 3E, F). Analysis of tumor-infiltrating lymphocytes (TIL) function, evaluated by GZMB and IFNγ expression, showed that cPLA2α downregulation did not affect the ability of CD8+ T cells to secrete granzyme B‌‌ (GZMB) (Fig. 3G) and interferon gamma (IFN-γ) (Supplementary Fig. 4A). As indicated by PD-1+ and Tim3+ staining, the exhaustion status of infiltrated CD8+ T cells was not significantly different between the two groups (Supplementary Fig. 4C, D). Consistent results were observed for CD4+ T cell (Supplementary Fig. 4B, E, F). These findings suggested that although cPLA2α knockdown increased the tumor-infiltrating CD8+ T cells, it did not further enhance their cytolytic capacity. As a subgroup of immunosuppressive T cells, Treg cells may promote tumor proliferation by facilitating immune evasion25. Our data showed that the proportion of Tregs in the cPLA2α downregulation group was significantly lower than that in the control group (Fig. 3H). NK cells are important innate immune cells that exert natural cytotoxic effects on tumor cells and produce cytokines to participate in the antitumor immune response26. As illustrated in Fig. 3I, cPLA2α downregulation significantly increased the proportion of NK cells. cPLA2α knockdown affects tumor-associated myeloid cells (TAMCs) We investigated the effects of cPLA2α inhibition on the composition of TAMCs and their immunosuppressive function in the tumor microenvironment by using a subcutaneous LLC tumor-bearing C57BL/6 mouse model. The proportion of DC cells (CD45+CD11b+CD11c+) did not change significantly (Fig. 4A). However, the proportion of neutrophils (CD45+MHCII−Gr1+CD11b+) was significantly reduced in the cPLA2α downregulation group (Fig. 4B). Although no significant difference was observed in the proportion of total macrophages (CD45+CD11b+F4/80+) between the two groups, the proportions of CD45+CD11b+F4/80+MHC-II+ proinflammatory M1 phenotype (anti-tumor) and CD45+CD11b+F4/80+CD206+ immunosuppressive M2 phenotype (pro-tumor) significantly increased and decreased, respectively. These findings indicated that cPLA2α downregulation promoted the polarization of macrophages from M2 to M1 (Fig. 4C).Fig. 4Knockdown of cPLA2α influences the infiltration of tumor-associated myeloid cells.Flow cytometric analysis of tumor-infiltrating myeloid cells from C57BL/6 mice model in sh-Ctrl and sh-cPLA2α groups. Representative flow analysis and quantification of DC cells (n = 5) (A), Neutrophil (n = 4–5) (B), Macrophage, M1 macrophage, M2 macrophage and M1/M2 ratio (n = 3) (C); Flow cytometric analysis of tumor-infiltrating immune cells from BALB/c Nude mice model in sh-Ctrl and sh-cPLA2α groups. Representative flow analysis and quantification of CD45+ immune cells (n = 5) (D), Neutrophil (n = 5) (E), Macrophage, M1 macrophage, M2 macrophage and M1/M2 ratio (n = 5) (F). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. Knockdown of cPLA2α influences the infiltration of tumor-associated myeloid cells. Flow cytometric analysis of tumor-infiltrating myeloid cells from C57BL/6 mice model in sh-Ctrl and sh-cPLA2α groups. Representative flow analysis and quantification of DC cells (n = 5) (A), Neutrophil (n = 4–5) (B), Macrophage, M1 macrophage, M2 macrophage and M1/M2 ratio (n = 3) (C); Flow cytometric analysis of tumor-infiltrating immune cells from BALB/c Nude mice model in sh-Ctrl and sh-cPLA2α groups. Representative flow analysis and quantification of CD45+ immune cells (n = 5) (D), Neutrophil (n = 5) (E), Macrophage, M1 macrophage, M2 macrophage and M1/M2 ratio (n = 5) (F). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. We further examined the proportions of TAMCs in an immune-deficient BALB/c Nude mouse model with transplanted tumors. The proportion of CD45+ immune cells decreased slightly in the cPLA2α downregulation group (Fig. 4D). Compared with those in the control vehicle group, the proportions of neutrophils and macrophages were reduced in the cPLA2α downregulation group (Fig. 4E, F). Consistent with the observations in C57BL/6 mice, cPLA2α downregulation increased the proportion of M1 type and decreased the proportion of M2 type. In summary, cPLA2α downregulation reshaped the composition of TAMCs in the tumor microenvironment. In particular, the proportion of DC cells remained relatively unchanged, and that of neutrophils was significantly reduced. Macrophages polarized from the M2 phenotype to the M1 phenotype. These findings further suggested that cPLA2α downregulation activated the anti-tumor immunity to some extent. cPLA2α downregulation promotes PD-L1 expression, and its overexpression suppresses PD-L1 in LUAD cells Our previous results showed that cPLA2α knockdown inhibited tumor progression in immunodeficient mice but failed to slow tumor growth in C57BL/6 immune-competent mice. These results suggested that an intact immune system might compromise the anti-tumor efficacy of cPLA2α downregulation, prompting us to delve deeply into the underlying mechanisms. Through a systematic literature review, we found a study published in Cancer Cell in 2020 reporting similar results. In particular, they found that ADORA1 deletion inhibited the growth of melanoma cells in vitro and in transplanted tumor models of immunodeficient mice. However, in immune-competent mice, ADORA1 deletion triggered immune escape by upregulating PD-L1, leading to malignant proliferation27. We speculated that cPLA2α downregulation might trigger immune escape via certain mechanisms, thereby diminishing its own anti-tumor efficacy. To investigate whether cPLA2α downregulation induces immune escape in LUAD, we performed TISIDB online analysis to explore the relationship between cPLA2α expression levels and immunosuppressive factors. The results showed that cPLA2α was inversely associated with the expression of most immune escape genes in LUAD (Supplementary Fig. 5A). To further support these findings, we correlated cPLA2α transcript levels with those of transcripts encoding immunosuppressive factors by using a publicly available dataset (TCGA) of LUAD biopsies. The samples with high cPLA2α expression exhibited a low expression of immune escape genes, including PD-L1 (CD274), PD-1 (PDCD1), LAG3, PD-L2 (PDCD1LG2), FOXP3, and CTLA4 (Supplementary Fig. 5B). Collectively, these data further supported our hypothesis that cPLA2α downregulation might trigger immune escape in LUAD, thereby attenuating its own anti-tumor effect. PD-L1 is widely studied in immune escape. Our analysis based on the TCGA database showed that cPLA2α expression had the highest negative correlation with PD-L1 (CD274), with a correlation coefficient value of −0.32. Therefore, we focused on PD-L1. We performed Western blotting on tumors isolated from immune-competent C57BL/6 mice and found that PD-L1 protein expression was significantly induced in the tumor region of the cPLA2α downregulation group (Fig. 5A). This finding supported our hypothesis. Furthermore, the cPLA2α downregulation-mediated upregulation of tumor PD-L1 was further validated in LUAD cell lines in vitro. Western blotting revealed that PD-L1 protein levels were significantly increased in the cPLA2α downregulation groups of LLC and A549 cells (Fig. 5B). cPLA2α downregulation also upregulated PD-L1 transcription levels (Fig. 5D, E). Flow cytometry on LLC and A549 cell lines revealed the same results (Fig. 5H, I). Together, these results further supported the notion that cPLA2α downregulation promoted PD-L1 expression.Fig. 5cPLA2α knockdown induces PD-L1 expression, while overexpression inhibits it in LUAD cells.A The protein level of cPLA2α and PD-L1 expression in established C57BL/6 mice xenograft model assessed by western blotting (n = 5). B Western blotting detected PD-L1 expression in LLC cells and A549 cells after downregulation of cPLA2α (n = 3). C PD-L1 expression was assessed in LLC and H1299 cells after upregulation of cPLA2α by western blotting (n = 3). D, E The PD-L1 expression was upregulated in sh-cPLA2α groups of LLC and A549 cells by RT-qPCR assay (n = 3). F, G RT-qPCR assay showing PD-L1 expression between vector and cPLA2α overexpression groups in LLC and H1299 cells (n = 3). H, I Flow cytometry detected the PD-L1 expression in LLC and A549 cells between sh-Ctrl and sh-cPLA2α groups (n = 3). J, K Flow cytometry assessed PD-L1 expression between vector and cPLA2α overexpression groups in LLC and H1299 cells (n = 3). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, *P < 0.05, **P < 0.01, ***P < 0.001. n indicates biologically independent samples. cPLA2α knockdown induces PD-L1 expression, while overexpression inhibits it in LUAD cells. A The protein level of cPLA2α and PD-L1 expression in established C57BL/6 mice xenograft model assessed by western blotting (n = 5). B Western blotting detected PD-L1 expression in LLC cells and A549 cells after downregulation of cPLA2α (n = 3). C PD-L1 expression was assessed in LLC and H1299 cells after upregulation of cPLA2α by western blotting (n = 3). D, E The PD-L1 expression was upregulated in sh-cPLA2α groups of LLC and A549 cells by RT-qPCR assay (n = 3). F, G RT-qPCR assay showing PD-L1 expression between vector and cPLA2α overexpression groups in LLC and H1299 cells (n = 3). H, I Flow cytometry detected the PD-L1 expression in LLC and A549 cells between sh-Ctrl and sh-cPLA2α groups (n = 3). J, K Flow cytometry assessed PD-L1 expression between vector and cPLA2α overexpression groups in LLC and H1299 cells (n = 3). Data are presented as mean ± SD. Statistical test: Two-tailed unpaired Student’s t test, *P < 0.05, **P < 0.01, ***P < 0.001. n indicates biologically independent samples. To further validate that cPLA2α could affect PD-L1 expression, we generated LUAD cell lines with stable cPLA2α overexpression and found that cPLA2α overexpression significantly inhibited PD-L1 protein expression in LLC and H1299 cells (Fig. 5C). In addition, PD-L1 mRNA expression was downregulated in cPLA2α-overexpressing LLC and H1299 cell lines (Fig. 5F, G). Flow cytometry was used to detect PD-L1 surface expression in these cell lines, yielding consistent results (Fig. 5J, K). To investigate the effect of cPLA2α overexpression on IFNγ-induced PD-L1 expression, we treated cPLA2α-overexpressing LLC cells and control cells with IFNγ, a known inducer of PD-L1. Western blotting confirmed that cPLA2α overexpression significantly attenuated the IFNγ-mediated elevation of PD-L1 protein levels in LLC cell lines in vitro (Supplementary Fig. 6A, B). This result was also observed in cPLA2α-overexpressing H1299 cell lines (Supplementary Fig. 6C). Collectively, these findings suggested that cPLA2α overexpression decreased PD-L1 expression in LUAD cells. cPLA2α downregulation induces PD-L1 expression by regulating the STAT3 signaling pathway To further elucidate the downstream molecular mechanism of cPLA2α in mediating PD-L1 expression in LUAD, we used Western blotting to investigate several cancer-related signaling pathways involved in cPLA2α, namely, MAPK/ERK, PI3K-AKT, and JAK-STAT3 signaling pathways16,28,29. We found that cPLA2α downregulation inhibited the phosphorylation levels of ERK and AKT in LLC cells without affecting the total ERK and AKT expression (Supplementary Fig. 7). We further evaluated the protein expression of the STAT3 pathway. Although cPLA2α knockdown increased the protein level of p-STAT3 in LLC cell line, the total STAT3 level remained unchanged (Fig. 6A). We confirmed this finding in A549 cell line, where cPLA2α downregulation similarly increased the STAT3 phosphorylation levels (Fig. 6B). Notably, the activation of these signaling pathways, including PI3K/AKT, MAPK, and STAT3, has been demonstrated to be involved in the upregulation of PD-L1 in tumor cells30–34. These results suggested that cPLA2α knockdown might reduce PD-L1 expression by activating the STAT3 pathway. To further confirm the relationship between cPLA2α and the STAT3 signaling pathway, we detected the phosphorylation level of STAT3 in cPLA2α overexpressing LUAD cells. cPLA2α overexpression significantly reduced the phosphorylation level of STAT3 in LLC and H1299 cells (Fig. 6C, D). These results indicated that cPLA2α upregulation inhibited STAT3 phosphorylation.Fig. 6cPLA2α regulates PD-L1 expression via the STAT3 signaling pathway.Western blotting analysis of the STAT3 signaling pathway activation after downregulation of cPLA2α expression in LLC (A) and A549 (B) cells (n = 3). Western blotting analysis of the STAT3 signaling pathway inhibition after upregulation of cPLA2α expression in LLC (C) and H1299 (D) cells (n = 3). E LLC sh-cPLA2α and A549 sh-cPLA2α cells were pretreated with STAT3 pathway inhibitors Stattic (5 μM) for 24 h. Western blotting analysis of the protein levels of p-STAT3 and total STAT3 protein (n = 3). F Western blotting analysis of cPLA2α, PD-L1 and STAT3 protein expression was performed in established C57BL/6 mice xenograft model (n = 5). Data are presented as mean ± SD. Two-tailed unpaired Student’s t test was used for statistical analysis in (A–D, F), one-way ANOVA was used for statistical analysis in (E), ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001. n indicates biologically independent samples. cPLA2α regulates PD-L1 expression via the STAT3 signaling pathway. Western blotting analysis of the STAT3 signaling pathway activation after downregulation of cPLA2α expression in LLC (A) and A549 (B) cells (n = 3). Western blotting analysis of the STAT3 signaling pathway inhibition after upregulation of cPLA2α expression in LLC (C) and H1299 (D) cells (n = 3). E LLC sh-cPLA2α and A549 sh-cPLA2α cells were pretreated with STAT3 pathway inhibitors Stattic (5 μM) for 24 h. Western blotting analysis of the protein levels of p-STAT3 and total STAT3 protein (n = 3). F Western blotting analysis of cPLA2α, PD-L1 and STAT3 protein expression was performed in established C57BL/6 mice xenograft model (n = 5). Data are presented as mean ± SD. Two-tailed unpaired Student’s t test was used for statistical analysis in (A–D, F), one-way ANOVA was used for statistical analysis in (E), ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001. n indicates biologically independent samples. To further investigate whether the STAT3 signaling pathway is involved in cPLA2α-mediated PD-L1 upregulation, we treated cPLA2α-downregulated LLC cells with the STAT3 pathway inhibitor Stattic. Stattic covalently binds to the SH2 domain of STAT3, thereby inhibiting its phosphorylation, dimerization, and nuclear translocation. This effectively suppresses STAT3 pathway activation35–37. Western blotting demonstrated that Stattic effectively inhibited the cPLA2α knockdown-induced activation of p-STAT3 in LLC cells. When the STAT3 signaling pathway was inhibited by Stattic, the cPLA2α downregulation-mediated upregulation of PD-L1 was also reversed. Similarly, treatment of cPLA2α-downregulated A549 cells with Stattic inhibited their PD-L1 and p-STAT3 expression (Fig. 6E). We subsequently examined the expression of cPLA2α and p-STAT3 in our established xenograft model. Western blotting was used to detect p-STAT3 expression in the sh-Ctrl and sh-cPLA2α groups, revealing that the phosphorylation level of STAT3 was significantly upregulated in the cPLA2α downregulation group (Fig. 6F). These results confirmed that cPLA2α downregulation induced PD-L1 upregulation by activating the STAT3 signaling pathway, thereby attenuating its own antitumor effect on LUAD. cPLA2α knockdown enhances the therapeutic efficacy of PD-1 mAb in the treatment of LUAD syngeneic mouse models Combination treatments targeting the PD-L1 pathway have recently been utilized to enhance the antitumor T cell immunity of patients with LUAD. Our previous results demonstrated that cPLA2α downregulation increased the infiltration of CD8+ T cell without altering their cytotoxic function, partly explaining why the anti-tumor effects of cPLA2α downregulation are comprised in immune competent mice. Given our in vitro evidence that cPLA2α knockdown induced PD-L1 expression, we hypothesized that combining PD-1 mAb to block the PD-1/PD-L1 signaling pathway may exert anti-tumor effects. To testify this hypothesis, we used C57BL/6 mice to establish an allograft LUAD mouse model. In brief, stable cPLA2α downregulation or control cells were inoculated subcutaneously into C57BL/6 mice. When the tumor size reached an average of 30–50 mm3, the sh-cPLA2α or sh-Ctrl LLC tumor-bearing mice were intraperitoneally injected with PD-1 mAb (10 mg/kg) or solvent. Treatments with anti-PD-1 or solvent were administered every 3 days for a total of four doses, and tumor growth was measured every other day (Fig. 7A). As shown in (Fig. 7B–D), cPLA2α downregulation or PD-1 mAb alone exerted only a slight tumor suppressive effect. No statistically significant difference from the control group was observed. Notably, compared with the other three treatment groups, combination therapy with cPLA2α downregulation and PD-1 mAb significantly reduced the tumor volume and weight without causing serious toxicity in terms of body weight change (Fig. 7E). These results indicated that the combination of cPLA2α knockdown and PD-1 mAb therapy exerted a significant therapeutic effect on the LLC LUAD model, effectively inhibiting tumor growth. The therapeutic efficacy of cotreatment with PD-1 mAb and sh-cPLA2α was superior to that of anti-PD-1 monotherapy. Therefore, these findings suggested that cPLA2α is a potential target to enhance the efficacy of PD-1 mAb therapy for the individualized and precise treatment of patients with LUAD.Fig. 7Combination of cPLA2α inhibition and PD-1 mAb effectively suppresses LUAD tumor growth in vivo and increases the CTL infiltration.A Schematic diagram illustrating the treatment protocol of cPLA2α downregulation and/or PD-1 mAb in LLC cells constructed mice model. LLC sh-Ctrl and sh-cPLA2α cells were subcutaneously injected into C57BL/6 mice. When the tumor volume reached ≈50 mm3, the mice were treated with PD-1 mAb (10 mg/kg, every 3 days, 4 dose in total) or solvent intraperitoneally (n = 5 mice per group). At the endpoint, tumor-infiltrating lymphocytes and tumor cells were isolated for analysis. B Representative images of subcutaneous LLC LUAD tumors after the indicated treatments. C Tumor growth curve at the indicated time points (n = 5). D Tumor weights at the endpoint after the indicated treatments (n = 5). E Body weights of tumor-bearing mice receiving the indicated treatments (n = 5). Flow cytometric analysis of tumor-infiltrating lymphocyte in C57BL/6 mice xenograft model treated as indicated above (n = 3–5). Representative flow analysis and quantification of CD45+ immune cells (n = 5) (F) CD3+ T cells (n = 5) (G), CD8+ T cells, CD4+ T cells (n = 5) (H), GZMB+CD8+ T cells (n = 3–5) (I). Data are presented as mean ± SD. Statistical test: one-way ANOVA, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. Combination of cPLA2α inhibition and PD-1 mAb effectively suppresses LUAD tumor growth in vivo and increases the CTL infiltration. A Schematic diagram illustrating the treatment protocol of cPLA2α downregulation and/or PD-1 mAb in LLC cells constructed mice model. LLC sh-Ctrl and sh-cPLA2α cells were subcutaneously injected into C57BL/6 mice. When the tumor volume reached ≈50 mm3, the mice were treated with PD-1 mAb (10 mg/kg, every 3 days, 4 dose in total) or solvent intraperitoneally (n = 5 mice per group). At the endpoint, tumor-infiltrating lymphocytes and tumor cells were isolated for analysis. B Representative images of subcutaneous LLC LUAD tumors after the indicated treatments. C Tumor growth curve at the indicated time points (n = 5). D Tumor weights at the endpoint after the indicated treatments (n = 5). E Body weights of tumor-bearing mice receiving the indicated treatments (n = 5). Flow cytometric analysis of tumor-infiltrating lymphocyte in C57BL/6 mice xenograft model treated as indicated above (n = 3–5). Representative flow analysis and quantification of CD45+ immune cells (n = 5) (F) CD3+ T cells (n = 5) (G), CD8+ T cells, CD4+ T cells (n = 5) (H), GZMB+CD8+ T cells (n = 3–5) (I). Data are presented as mean ± SD. Statistical test: one-way ANOVA, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. To elucidate the alterations in immune features associated with the enhanced therapeutic efficacy of the combination therapy of cPLA2α knockdown and PD-1 mAb, we investigated the tumor immune microenvironment in LLC tumor-bearing mice treated with anti-PD-1 and cPLA2α knockdown, either in combination or alone. Flow cytometry results showed there was no significant difference in the total number of CD45+ immune cells among the four treatment groups (Fig. 7F). The number of CD3+ T cells increased in the three treatment groups compared with that in the control group (Fig. 7G). The proportion of tumor-infiltrating CD8+ T cells was significantly higher in the sh-cPLA2α + PD-1 mAb group than in the other three groups, and no significant changes were observed for CD4+ T cells (Fig. 7H). More importantly, the percentage of GZMB+CD8+ T cells substantially increased in the sh-cPLA2α + PD-1 mAb group (Fig. 7I), indicating the enhanced activation of cytotoxic T effector cells. These results suggested the synergistic benefit of cPLA2α downregulation and PD-1 antibody treatment on cytotoxic granule secretion. No significant difference in the ability of CD8+ T cells and CD4+ T cells to secrete INFγ was found among the four groups (Supplementary Fig. 8A, B). We also detected the PD-1 levels of CD8+ T cells and CD4+ T cells and found that they were significantly reduced in the control group and cPLA2α knockdown group after treatment with PD-1 antibody (Supplementary Fig. 8C, D). Meanwhile, PD-1 antibody substantially improved the infiltration of Tim3+CD8+ T cells in the control group and cPLA2α knockdown group (Supplementary Fig. 8E). No significant changes in the infiltration of Tim3+CD4+ T cells were observed among the treatment groups (Supplementary Fig. 8F). These findings provided evidence that PD-1 antibody could ameliorate cPLA2α knockdown-induced tumor immunosuppression, and cPLA2α downregulation enhances the therapeutic sensitivity to PD-1 mAb. Overall, these results suggested that the combination of cPLA2α knockdown and PD-1 mAb treatment promoted the infiltration and functional activation of CD8+ T cells. The tumor microenvironment is a complex structure, and immunosuppression within it is primarily mediated by neutrophils and tumor-associated macrophages (TAMs). In our previous studies, we observed significant changes in the composition of TAMCs between cPLA2α knockdown and control groups. Therefore, we next examined the proportion of myeloid cells subsets in tumor samples from the four treatment groups. Compared with that in the control group, the proportion of infiltrated neutrophils in the tumor was reduced in the three treatment groups, with the most significant suppression observed in the sh-cPLA2α + PD-1 mAb group (Fig. 8A). In addition, we measured the proportion of macrophages. The total macrophage infiltration was significantly inhibited in the three treatment groups compared with that in the control group, with the most pronounced effect observed in the sh-cPLA2α + PD-1 mAb group (Fig. 8B). The combination treatment further increased the proportion of inflammatory M1-like macrophages and decreased the proportion of immunosuppressive M2-like macrophages, as evidenced by the increased M1/M2 ratio (Fig. 8C, D). Finally, we detected PD-L1 expression among the four treatment groups in the allograft tumor at day 22. As depicted in Fig. 8E, F, PD-L1 protein was upregulated in the sh-cPLA2α group and sh-cPLA2α + PD-1 mAb group, further validating that cPLA2α knockdown could promote PD-L1 expression during allograft model assay. In summary, the combination of cPLA2α downregulation and PD-1 monoclonal antibody therapy altered the composition of myeloid immune cells in tumors, characterized by a significant decrease in the proportion of neutrophils and a substantial increase in the polarization of macrophages from the M2-like to the M1-like state.Fig. 8Combination of cPLA2α inhibition and PD-1 mAb affects tumor-associated myeloid cells.Flow cytometric analysis of tumor-infiltrating myeloid cells and tumor cells in C57BL/6 mice xenograft model treated as indicated above (n = 3–5). Representative flow analysis and quantification of Neutrophil (n = 4–5) (A), Macrophage, M1 macrophage, M2 macrophage, M1/M2 ratio (n = 3–5) (B–D), tumor cells (n = 5) (E) and PD-L1 membrane expression of tumor cells (n = 5) (F). Data are presented as mean ± SD. Statistical test: one-way ANOVA, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. Combination of cPLA2α inhibition and PD-1 mAb affects tumor-associated myeloid cells. Flow cytometric analysis of tumor-infiltrating myeloid cells and tumor cells in C57BL/6 mice xenograft model treated as indicated above (n = 3–5). Representative flow analysis and quantification of Neutrophil (n = 4–5) (A), Macrophage, M1 macrophage, M2 macrophage, M1/M2 ratio (n = 3–5) (B–D), tumor cells (n = 5) (E) and PD-L1 membrane expression of tumor cells (n = 5) (F). Data are presented as mean ± SD. Statistical test: one-way ANOVA, ns not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n indicates biologically independent samples. Discussion The potential role of cPLA2α in the tumor immune microenvironment of LUAD, particularly its involvement in tumor immune escape, remains poorly understood. In addition, the interaction between cPLA2α knockdown and ICIs therapy is not well elucidated. This study demonstrated that cPLA2α downregulation reduced LUAD growth in immunodeficient mice but not in immune competent mice. Further investigation indicated that cPLA2α knockdown increased the proportion of infiltrating CD8+ T cells and decreased the percentage of immunosuppressive myeloid cells. Through in vitro and in vivo studies, we mechanistically demonstrated that cPLA2α knockdown induced PD-L1 upregulation via the activation of STAT3 phosphorylation, thereby weakening the anti-tumor effects. In a preclinical experiment, we showed that cPLA2α knockdown enhanced the efficacy of PD-1 mAb in a LUAD mouse xenograft model. Therefore, cPLA2α is a potential therapeutic target for LUAD, capable of enhancing the efficacy of ICIs for LUAD treatment (Fig. 9).Fig. 9Schematic of the mechanism of cPLA2α inhibition enhances PD-1 mAb efficacy by optimizing the tumor immune microenvironment in lung adenocarcinoma.cPLA2α knockdown plays dual and context-dependent roles in LUAD. It directly suppresses tumor growth by inhibiting malignant proliferation in vitro and in xenograft models in immunodeficient mice, whereas in immunocompetent settings it significantly altered the tumor immune microenvironment. This intervention enhances immunostimulatory components by elevating CD8+ T cell and M1 macrophage infiltration, while reducing immunosuppressive cells including Treg cells, M2 macrophages, and neutrophils. Paradoxically, cPLA2α knockdown induces PD-L1 expression via STAT3 signaling activation, partially counteracts the tumor-suppressive effect of cPLA2α depletion. Preclinically, inhibition of cPLA2α synergizes with anti-PD-1 therapy by increasing cytotoxic T lymphocyte infiltration, suggesting a viable combinatorial immunotherapeutic strategy for LUAD patients. Schematic of the mechanism of cPLA2α inhibition enhances PD-1 mAb efficacy by optimizing the tumor immune microenvironment in lung adenocarcinoma. cPLA2α knockdown plays dual and context-dependent roles in LUAD. It directly suppresses tumor growth by inhibiting malignant proliferation in vitro and in xenograft models in immunodeficient mice, whereas in immunocompetent settings it significantly altered the tumor immune microenvironment. This intervention enhances immunostimulatory components by elevating CD8+ T cell and M1 macrophage infiltration, while reducing immunosuppressive cells including Treg cells, M2 macrophages, and neutrophils. Paradoxically, cPLA2α knockdown induces PD-L1 expression via STAT3 signaling activation, partially counteracts the tumor-suppressive effect of cPLA2α depletion. Preclinically, inhibition of cPLA2α synergizes with anti-PD-1 therapy by increasing cytotoxic T lymphocyte infiltration, suggesting a viable combinatorial immunotherapeutic strategy for LUAD patients. Chronic inflammation plays a significant role in the development and progression of cancer. The dynamic balance between immune surveillance and protumorigenic inflammation constitutes the core feature of the tumor immune microenvironment38. cPLA2α functions in inflammation and tumor promotion through the arachidonic acid cascade pathway14. The cPLA2α–COX2–PGE2 axis is a key pathway involved in mediating immune suppression and tumor promotion23,39. It has been demonstrated that targeting COX2/PGE2 restored the antitumor immunity by remodeling the tumor microenvironment, which increased the efficacy of ICIs in melanoma and NSCLC mouse models40,41. However, the role of the upstream regulatory factor cPLA2α in tumor immunity, especially in LUAD, poorly understood. Previous research on cPLA2α focused on its direct effects on tumor cells15–17; its influence on host immunity and tumor microenvironment is poorly understood. In recent years, its role in tumor immunity has gained increasing academic attention. Kong et al.20 found that the abnormally activated cPLA2α–arachidonic acid pathway in colorectal cancer led to CD39+ γδ Treg cell polarization, consequently enhancing immunosuppression and colorectal cancer progression. Ramesh et al.21 demonstrated that targeting cPLA2α reduced the malignancy of breast cancer by suppressing the recruitment of immunosuppressive myeloid cells and increasing the infiltration of CD8+ T cells. However, the role of cPLA2α in the efficacy of ICIs has not yet been explored. Our study further characterized the immunomodulatory capacity of cPLA2α in LUAD, focusing on its systemic role in immunotherapy. Here, we observed that cPLA2α downregulation inhibited tumor growth in the transplanted tumor models of immunodeficient mice, consistent with in vitro observations. Basing on the results of bioinformatics analysis, we anticipated that the effect of cPLA2α knockdown on tumor proliferation would be highly pronounced in an allograft tumor model with an intact immune system. Surprisingly, no significant differences in tumor growth and volume were observed. Recent works have investigated the role of cPLA2α in modulating the tumor microenvironment by altering the recruitment of immunosuppressive cells. However, only a few studies have explored the effects of cPLA2α on the overall immune landscape, particularly in LUAD. In our present study, cPLA2α downregulation reshaped the tumor immune microenvironment of LUAD by promoting the infiltration of CD8+ T cells, increasing NK cells, and decreasing Treg cells. An elevated peripheral blood neutrophil to lymphocyte ratio predicts poor prognosis in various cancers, including lung cancer42,43. In the tumor microenvironment, neutrophils can inhibit tumor progression through cytotoxicity and the activation of other immune cells44,45. However, neutrophils also participate in tumor occurrence and development by promoting angiogenesis, regulating tumor survival and metastasis, and modulating immune response46,47. We observed that the proportion of neutrophils was significantly reduced in the cPLA2α downregulation group. TAMs are the most important immunosuppressive cells in the tumor microenvironment, mediating immunosuppression by synthesizing and secreting anti-inflammatory factors and thereby promoting tumor growth and progression48. Macrophages can be divided into two phenotypes: type M1 inhibits tumor growth, and type M2 promotes tumor growth. Targeting TAMs and promoting the phenotypic transformation of M2-type to M1-type macrophages have recently attracted increasing attention as anti-cancer strategies49–51. We discovered that cPLA2α knockdown strongly promoted the polarization of macrophages from the M2-type to the M1-type. These results confirmed our previous database findings that high cPLA2α expression was associated with poor immune infiltration in LUAD. The PD-L1/PD-1 signaling pathway, which leads to tumor immune escape, remains a research hotspot and a challenging area. Therefore, further investigating the regulatory mechanisms of PD-L1 expression is essential to prevent tumor immune evasion and develop effective immunotherapy strategies. In this study, we observed that an intact immune system compromised the anti-tumor efficacy of cPLA2α downregulation in the LLC LUAD tumor model. Through literature review and further database analysis, we demonstrated that cPLA2α knockdown significantly increased PD-L1 mRNA and protein levels in LUAD cell lines, and its overexpression showed the opposite effect. This finding was also validated in vivo. Therefore, our results suggested that tumor PD-L1 might contribute to the unsatisfied anti-tumor effect of cPLA2α knockdown in the immunocompetent mice model. Whether the switch between tumor-inhibition and tumor-promotion effects of cPLA2α depends on PD-L1 must be further clarified through in vivo and in vitro experiments. The regulation of PD-L1 expression involves chromatin and genome changes, transcriptional and posttranscriptional regulation, and translational and posttranslational modification. In addition, multiple factors in the tumor microenvironment can induce the expression of tumor PD-L111,52. Several classical tumor signaling pathways, such as PI3K/AKT, MAPK, and STAT3, have been confirmed to be involved in the regulation of PD-L1 expression30–34. Our in vitro and in vivo studies indicated that the activation of STAT3 pathway might be involved in the cPLA2α downregulation-induced upregulation of PD-L1. In our study, we suggested that cPLA2α knockdown activated STAT3 and thereby induced PD-L1 expression and promoted immune escape in LUAD. These findings further expanded our understanding of the role of cPLA2α, and even the entire PLA2 family, in tumor immunity. Interestingly, previous studies reported that PEG2, a primary product of cPLA2α, promotes the progression of cholangiocarcinoma and prostate cancer by activating STAT329,53. However, our study showed that cPLA2α downregulation promoted the phosphorylation of STAT3, a function that has never been reported for the PLA2 family. Given the extensive involvement and complexity of cPLA2α and its downstream products in a variety of physiological and pathological processes and its complex dual role in different tumors, cPLA2α possibly mediates intricate signal transduction and crosstalk in the tumor microenvironment and exhibits different regulations of STAT3 activity. Although our data suggested that the cPLA2α downregulation-induced upregulation of PD-L1 is mediated by STAT3 activation, the exact molecular mechanism still requires further exploration. ICIs represented by PD-1 or PD-L1 antibodies have achieved major breakthroughs in the treatment of various cancer types. However, a significant number of patients exhibit primary resistance or develop acquired resistance after relapse2. Therefore, it is necessary to combine ICIs with other therapeutic strategies to further improve the treatment efficacy. Our results demonstrated that cPLA2α downregulation improved the anti-tumor efficacy of PD-1 mAb. We also showed that their combination resulted in the extensive infiltration of GZMB+CD8+ T cells. Although cPLA2α knockdown alone increased CD8+ T cell infiltration, it was not sufficient to initiate antitumor activity. This finding further emphasized the importance of cPLA2α as a target to enhance immunotherapy efficacy. Given the complexity and intricate mechanisms of the interaction between tumor and immune cells, the increased secretion of GZMB by CD8+ T cells induced by the combination therapy might be one mechanism to restore the immune response. Another possible reason was the cPLA2α knockdown-induced upregulation of PD-L1. Although the exact mechanisms must be further explored, these findings indicated that the PD-L1 expression tends to be high for patients with LUAD exhibiting low cPLA2α expression. Blocking the PD-1/PD-L1 pathway by combining PD-1/PD-L1 mAb might offer potential benefits, which must be validated in a cohort of patients with LUAD treated with immunotherapy. In a sense, we identified potential screening strategies for assessing the efficacy of PD-1 mAb therapy in patients with LUAD. While our findings highlight cPLA2α as a compelling therapeutic target in LUAD, translating this concept faces several challenges54–56. First, achieving high inhibitor specificity is crucial to avoid disrupting the physiological functions of other PLA2 isoforms. Second, potential on-target toxicity must be evaluated, given cPLA2α’s role in normal physiological processes. Third, an incomplete understanding of isoform-specific roles in different cancers limits its antitumor efficacy. Our research findings have revealed the specific tumor promotion and immunomodulatory functions of cPLA2α in LUAD. This upstream strategy may offer broader efficacy than downstream COX-2 blockade but concurrently faces the existing challenges of functional redundancy and potential toxicity. Consequently, future translation should develop selective cPLA2α inhibitors and explore rational combination therapies to overcome compensatory pathways. Nevertheless, we must acknowledge that this part of the study still has some limitations. First, our current study only demonstrates that cPLA2α downregulation is a promising option to enhance the therapeutic efficacy of PD-1 mAb. Future work is needed to explore the potential synergism of cPLA2α inhibitors with other ICIs, such as anti-CTLA4 and anti-LAG3, or with cancer vaccines. Moreover, the potential molecular pathways that establish cellular links or crosstalk between cancer and immune cells and the intercellular connections among immune cells require further investigation. Additionally, the precise mechanism by which cPLA2α drives STAT3 activation, whether mediated through its bioactive lipid metabolites, direct protein–protein interactions, or alternative molecular pathways, is critical to comprehensively decipher its dual functional roles in modulating inflammatory signaling and immune evasion phenotypes. This mechanistic clarity holds significant clinical transformation implications for optimizing combinatorial immunotherapy strategies. Our study revealed the important role of cPLA2α in the tumor microenvironment of LUAD. On the one hand, cPLA2α downregulation increased the infiltration of CD8+ T cells and decreased the percentage of immunosuppressive cells. On the other hand, cPLA2α regulated PD-L1 expression via the STAT3 signaling pathway, thereby attenuating the anti-tumor effect of cPLA2α downregulation on LUAD. Furthermore, knocking down cPLA2α enhanced the efficacy of PD-1 mAb by increasing the infiltration of cytotoxic T cells. These results demonstrated that patients with LUAD showing a high cPLA2α expression and a low PD-L1 expression may potentially benefit from cPLA2α inhibitor therapy, which suppresses tumor proliferation and enhances the efficacy of PD-1 mAb treatment via upregulating tumor PD-L1. Although most cPLA2α inhibitors have failed in clinical trials due to their poor efficacy and adverse effects, the development of novel agents targeting cPLA2α is necessary because of its potential as an important anti-tumor agent and to synergistically enhance the efficacy of immunotherapy. Methods Patients and specimens A total of 182 human LUAD tissue samples were retrospectively identified at Tianjin Medical University Cancer Institute and Hospital (Tianjin, China) between May 2013 and November 2017. All the tissue samples were verified as LUAD by two independent pathologists and were staged in accordance with the eighth edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual. This study was conducted in line with the ethical guidelines of the Helsinki Declaration and was approved by the Ethics Committee. Bioinformatics analysis Gene Expression Profiling Interactive Analysis 2 (GEPIA2) and University of Alabama at Birmingham Cancer Data Analysis Portal (UALCAN) databases were used to analyze the mRNA and protein expression levels of PLA2G4A in LUAD and normal tissues. The RNA-seq expression data (in TPM format) and corresponding clinical information of the lung adenocarcinoma (LUAD) cohort were obtained from The Cancer Genome Atlas (TCGA) via the GDC data portal (https://portal.gdc.cancer.gov). The survival analysis was performed to evaluate the prognostic value of PLA2G4A utilizing data from TCGA-LUAD data and employing R packages. The relationship between PLA2G4A expression and infiltrating immune cells, chemokines, chemokine receptors, immunostimulators, immunoinhibitors, and MHC molecules was analyzed using the Tumor-Immune System Interaction Database (TISIDB) and R Studio. Immunohistochemistry (IHC) LUAD tissue sections were sequentially immersed in xylene and gradient ethanol to deparaffinize and rehydrate. Antigen repair was performed in citric acid buffers, followed by the inhibition of endogenous peroxide activities with 3% hydrogen peroxide for 10 min. The samples were incubated with rabbit anti-PLA2G4A (1:100, GeneTex, GTX110218; Irvine, USA) for 30 min at room temperature and overnight at 4 °C. On the 2nd day, the samples were washed three times, incubated with a secondary antibody for 1 h at room temperature, visualized with 3, 3-diaminobenzidine solution, and counterstained with hematoxylin. The staining intensity was evaluated in four grades (0, no immune response; 1, weak immune response; 2, moderate immune response; and 3, strong immune response). The percentage of positive cells was scored from 0 to 3 (0, no positive cells; 1, <30% positive cells; 2, 30–60% positive cells; and 3, 60–100% positive cells). The final IHC score was the product of the staining intensity score and the percentage score and ranged from 0 to 9 (0, negative; 1–2, low expression; 3–4, moderate expression; and 6–9, high expression). Western blotting and antibodies Freshly isolated LUAD and adjacent nonmalignant tissue samples were frozen in liquid nitrogen after surgery, and the tissue proteins were extracted using a tissue protein extractor (Servicebio, China). Furthermore, the cells were lysed on ice for 30 min in 1 × SDS lysis buffer supplemented with 1 mM NaF, 1 mM Na3VO4, and a 1 × protease/phosphatase inhibitor cocktail (Roche, Switzerland). The collected protein was denatured at 95 °C for 10 min and centrifuged at 12,000 rpm and 4 °C for 10 min. Then the sample was quantified by NanoDrop2000 spectrophotometer. Equal amounts of protein were loaded onto a gel and separated by SDS-PAGE. The protein was then transferred to a PVDF membrane (Immobilon-P; Millipore, USA) and blocked with 5% bovine serum albumin for 2–4 h. The membranes were incubated with primary antibodies overnight at 4 °C and secondary antibody for 1 h at room temperature. Protein molecular weights are indicated in kilodaltons (kDa). The following antibodies were used: anti-cPLA2α (1:500; GTX110218) from Gene Tex (Irvine, USA), anti-PD-L1-mouse (1:1000) from Abcam (MA, USA; ab213480), anti-PD-L1-human (1:1000; T55980S) from Abmart (Shanghai, China), anti-p-STAT3 (1:1000; 9245S) and anti-STAT3 (1:1000; 4904S) from Cell Signaling Technology (Beverly, USA), and anti-GAPDH (1:500; sc-47724) from Santa Cruz Biotechnology (Santa Cruz, USA). RNA extraction, cDNA synthesis, and quantitative real-time PCR (qRT-PCR) Total RNA was extracted from cells using TRIzol reagent (Ambion, USA). cDNA was synthesized by RNA reverse transcription using the qRT-PCR kit (Takara, Japan). Amplification was conducted using the AceQ qPCR SYBR Green Master Mix kit in accordance with the manufacturer’s instructions (Vazyme, China). Each sample was analyzed at least three biological replicates and normalized to the endogenous reference control gene β-actin. Relative gene expression levels were calculated using the 2−ΔΔCt method. Primer sequences were designed using the NCBI Primer-BLAST tool and synthesized by Beijing Genomics Institute. The primer sequences are listed in Supplementary Data 3. Cell culture Lewis lung carcinoma (LLC), A549, and H1299 cells were purchased from American Type Culture Collection (ATCC, Manassas, VI). The LLC cells were cultivated in DMEM medium (Corning, USA), and the A549 and H1299 cells were cultivated in 1640 medium (Corning, NY, USA); both were supplemented with 10% fetal bovine serum (FBS; PAN-Seratech, Germany) and 1% penicillin/streptomycin (PS; HyClone, USA) under culture requirements (37 °C and 5% CO2). Cell transfection Packaging plasmids (VSVG and ΔR) and expression plasmids (sh-Ctrl, sh-cPLA2α, Vector, and cPLA2α) were transfected into HEK293T cells using Lipofectamine 2000 (Invitrogen, USA) to obtain lentiviral particles. Lentiviruses were then produced by the HEK293T cells. LLC cells were infected with a lentivirus to produce stable cPLA2α knockdown (KD) and cPLA2α overexpression (OE) cells and matched control (Ctrl) or vector cell lines under puromycin (Gibco, USA) selection. Stable cPLA2α KD and OE cells were also produced from A549 and H1299 cells using the same method. Cell viability assay The cells were plated in a 96-well plate at 1 × 103 cells/well. Five parallel wells were set up for each group, separately added with 10 μL of CCK-8 reagent (Biosharp, China), and incubated with the cells in a 37 °C incubator for 1 h. Optical density (OD) value at 450 nm was measured using an enzyme labeling instrument. A cell proliferation curve was drawn after continuous detection on days 0, 1, 2, 3, and 4. Chemotaxis assay A total of 1 × 105 cells in FBS-free medium was added to the upper cell chamber of a Transwell system with 8 μm pore-size polyethylene membranes, and 600 μL of medium containing 20% FBS was loaded to the lower chamber. After incubation for 12–18 h at 37 °C with 5% CO2, the cells on the bottom were fixed with 4% paraformaldehyde, stained with 0.1% crystal violet, and counted under a microscope. Animal models and in vivo treatment BALB/c Nude mice and C57BL/6 mice (males, 4–5 weeks of age) were purchased from SPF Biotechnology (Beijing, China). Mice were housed in individually ventilated cages (5 mice per cage) under controlled conditions (temperature of 22 ± 2 °C, relative humidity of 50 ± 10%, and a light/dark cycle of 12 h). The LLC sh-Ctrl and sh-cPLA2α cells (5 × 105) in 100 μL of PBS were subcutaneously injected into the groin of these mice. For in vivo treatment, the mice were randomly allocated into four groups (n = 5 per group) when the tumor volumes reached approximately 30–50 mm3: sh-Ctrl (solvent alone), sh-Ctrl + PD-1 monoclonal antibody (mAb), sh-cPLA2α (solvent alone), and sh-cPLA2α + PD-1 mAb. PD-1 antibody (10 mg/kg) or solvent was injected intraperitoneally every 3 days. Tumor volume and animal weight were measured every 2–3 days. Tumor growth was analyzed by measuring tumor length and width, and tumor volume was calculated as volume (mm3) = length × width2 /2. Mice were euthanized by CO₂ inhalation at weeks 3–4, when the tumor diameter had not yet exceeded the maximum allowed limit of 20 mm, and their tumor weights were recorded. Half of the harvested fresh tumor tissues were frozen in liquid nitrogen for subsequent Western blotting, and the other half was used for flow cytometry. The maximal measured size of the subcutaneous tumors was 20 mm in diameter in accordance with the animal ethics committee. In none of the animal experiments was this limit exceeded. The animal study protocol was approved by the Animal Ethics Committee of Tianjin Medical University Cancer Institute and Hospital. We have complied with all relevant ethical regulations for animal use. Flow cytometry Tumor tissues were separated, cut into small fragments, and digested for 1 h using the Tumor Dissociation Kit for mouse (Miltenyi, Germany) at 37 °C on a constant temperature shaker. Spleens were grinded in the copper mesh and then filtered through 70 and 40 μm cell strainers to obtain a single cell suspension. The cell suspension was lysed with ACK lysis buffer on ice for 5 min to lyse the red blood cells. Afterward, 2 × 106 cells were incubated with fixable viability dye (Biolegend, 1:1000) for 30 min at 4 °C and then with anti-mouse Fc-block CD16/32 antibody (clone 93, 1:200, Biolegend, 101320; San Diego, USA) under the same conditions. Finally, the cell membrane proteins were stained with antibodies for 40 min at 4 °C. For intracellular staining, the cells were fixed, permeabilized, and incubated with antibodies. For cytokine staining, the cells were incubated for 5 h with cell activation cocktail (1:500, Biolegend, 423303; San Diego, USA) at 37 °C, supplemented with DMEM medium containing 10% FBS, and stained as previously described. Data were acquired using a FACS Aria flow cytometer (BD) with CellQuest software and analyzed with FlowJo software. The antibodies were as follows: anti-TNF-α (clone MP6-XT22, 1:100, 506304), anti-IFN-γ (clone XMG1.2, 1:100, 505808), anti-GZMB (clone QA16A02, 1:100, 372206), anti-Tim3 (clone RMT3-23, 1:200, 119703), anti-CD274 (clone 10 F.9G2, 1:200, 124333), anti-CD11b (clone M1/70, 1:200, 101211), and anti-Ki67 (clone 11F6, 1:100, 151210) from Biolegend (San Diego, USA); anti-CD45 (clone 30-F11, 1:200, 552848), anti-CD4 (clone RM4-5, 1:200, 553051), anti-NK1.1 (clone PK136, 1:200, 550627), and anti-Ly6G/Ly6C (clone RB6-8C5, 1:200, 553128) from BD Biosciences (Franklin Lakes, USA); and anti-CD3 (clone 17A2, 1:200, 48-0032-82), anti-CD8 (clone 53-6.7, 1:200, 45-0081-82), anti-CD279 (clone J43, 1:200, 11-9985-81), anti-F4/80 (clone BM8, 1:200, 25-4801-8), anti-Foxp3 (clone FJK-16s, 1:100, 12-5773-82), anti-CD206 (clone MR6F3, 1:200, 12-2061-82), and anti-MHC-II (clone M5/114.15.2, 1:200, 11-5321-82) from eBioscience. Calculation of M1/M2 ratio The main gating strategies for flow cytometry data analysis are shown in Supplementary Fig. 3. Specially, the M1/M2 ratio was derived directly from flow cytometry data of tumor infiltrating immune cells and was calculated as follows: The gating strategy was: Live CD45+ leukocytes → CD11b+ F4/80+ macrophage population → M1 (CD11b+F4/80+MHCII+) and M2 (CD11b+F4/80+ CD206+) subsets. The M1/M2 ratio was calculated using the formula: M1/M2 ratio = (the percentage of M1 macrophages)/(the percentage of M2 macrophages). Drugs and reagents Murine PD-1 antibody (clone RMP1-14) was purchased from Bio X Cell. Stattic was obtained from MedChemExpress (MCE, USA). IFNγ was purchased from ABclonal (Wuhan, China). Statistics and reproducibility ‌The statistical analyses in this study were performed by R (version 4.5.1) and GraphPad Prism 10. Each experiment involves at least three biological replicates. All biological experiment data was presented as mean ± standard error (SD). The Shapiro-Wilk test and Bartlett’s test were used to evaluate the normality hypothesis and the homogeneity hypothesis of variance. Two independent samples t-tests were used to compare the means of two groups, and one-way analysis of variance was performed on three or more groups. When the homogeneity of variance assumption cannot be met, Welch’s correction was applied. Tukey’s or Dunnett’ T3 multiple comparisons test were chosen for Post Hoc. The chi-square test was used to compare sample rates. The survival curves were described by Kaplan–Meier method, where the difference was acquired by the log-rank test, and univariate and multivariate Cox analysis was applied to explore the hazard ratio of each factor. P < 0.05 was considered statistically significant. Ethics approval and consent to participate This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Tianjin Medical University Cancer Institute and Hospital. And all participants signed informed consent statements. All ethical regulations relevant to human research participants were followed. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Transparent Peer Review file Supplementary Information Description of Additional Supplementary Files Supplementary Data 1 Supplementary Data 2 Supplementary Data 3 Supplementary Data 4 Reporting summary Transparent Peer Review file Supplementary Information Description of Additional Supplementary Files Supplementary Data 1 Supplementary Data 2 Supplementary Data 3 Supplementary Data 4 Reporting summary