Abstract Background Osteosarcoma (OS) lung metastases remain a significant therapeutic challenge. Innate immune activation is a promising therapeutic approach. Innate immune agonists can modulate the tumor immune microenvironment and improve therapeutic response. Methods Using an experimental syngeneic OS lung metastasis BALB/c mouse model with K7M3-luc OS cells, we evaluated the antitumor effects of yeast-derived particulate β-glucan in prevention and therapeutic settings. We then assessed whether the CD40 agonist (CD40a) in combination with β-glucan increased therapeutic response in two different immune-competent mouse models of OS lung tumor burden. Results In the pretreatment settings, mice treated with β-glucan prior to OS cell infusion developed significantly fewer lung tumor burdens and had increased survival. Pretreatment with β-glucan prevented tumor cell seeding in the lungs. In tumor-bearing mice, β-glucan treatment significantly suppressed tumor growth and prolonged overall survival. β-glucan treatment increased activated pro-inflammatory M1-like macrophages and natural killer (NK) cells secreting interferon-γ and granzyme B in the lungs. Depletion studies showed that the antitumor effect of β-glucan was dependent on macrophages and NK cells. Additionally, β-glucan treatment also induced myelopoiesis in the bone marrow. The therapeutic benefit of β-glucan was further augmented when combined with CD40a. Combination therapy significantly increased the infiltration of activated macrophages, including tumor necrosis factor-α secreting macrophages, and NK cells into the lungs compared with monotherapy. Bulk RNA sequencing of lung tissue revealed that the combination treatment group exhibited enhanced activation of antitumor innate immune pathways. Conclusions Collectively, our findings demonstrate the antitumor activity of β-glucan against OS lung tumor burden, that combining β-glucan and CD40a increases therapeutic activity, and that this activity is mediated by activation of innate immunity (macrophages and NK cells). Keywords: Immunotherapy, Solid tumor, Innate, Immune modulatory __________________________________________________________________ WHAT IS ALREADY KNOWN ON THIS TOPIC * Patients with relapsed or refractory osteosarcoma lung metastases have poor overall survival outcomes, which have remained stagnant for decades. Despite advances in cancer immunotherapy, no meaningful clinical success has been observed. WHAT THIS STUDY ADDS * The study shows that β-glucan treatment elicits a potent antitumor response in both preventive and therapeutic models of osteosarcoma, leading to prolonged survival. This effect is associated with activation of macrophages and natural killer cells. Moreover, the addition of a CD40 agonist further enhances innate immune responses, amplifies antitumor immunity, and prolongs survival compared with monotherapy. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY * These preclinical data provide proof of concept for a novel immunotherapeutic strategy that harnesses innate immunity to treat osteosarcoma lung metastases. Background Osteosarcoma (OS) is a malignant bone tumor that predominantly presents in children and young adults. It is estimated that approximately 90% of newly diagnosed patients will have microscopic lung metastases at the time of diagnosis.[29]^1 2 The standard treatment includes surgical resection along with presurgical and postsurgical combination chemotherapy. Despite these aggressive treatments, the 5-year survival rate for patients with localized OS is about 60% but only 20% for patients with metastatic or recurrent disease.[30]^1 These survival rates have not changed for >25 years, which underscores the critical need for novel therapeutic approaches, especially for patients with metastatic OS. Cancer immunotherapy, including immune checkpoint inhibitors, against OS lung metastases has not demonstrated clinical success. This therapeutic resistance is in part due to an immunosuppressive tumor microenvironment (TME) characterized by infiltration of myeloid-derived suppressor cells, and immune-suppressive M2 macrophages, which inhibit the antitumor activity of T cells and natural killer (NK) cells. Furthermore, the infiltration of effector T cells into the tumor is limited with the majority of cells remaining on the periphery of the tumor.[31]3,[32]5 Collectively, this decreases the efficacy of immune therapies that rely on T-cell activity. Innate immune cells, such as macrophages and NK cells, play critical roles in antitumor immunity. Pulmonary OS metastases are frequently infiltrated by tumor-associated macrophages, a heterogeneous population ranging from immune-suppressive M2 (alternatively activated) macrophages to pro-inflammatory M1 (classically activated) macrophages. M2 macrophages secrete immunosuppressive cytokines, including interleukin (IL)-10 and Transforming gorwth factor beta (TGF-β), and their high infiltration correlates with poor prognosis in OS.[33]6,[34]8 In contrast, M1 macrophages produce pro-inflammatory cytokines such as IL-12 and tumor necrosis factor (TNF)-α, exhibit potent antitumor activity, and enhance NK and T cell effector functions. M1 macrophage infiltration in OS lung metastases is positively correlated with favorable patient outcome.[35]^6 Therefore, strategies to activate innate immune cells, particularly macrophages, have the potential to overcome myeloid-driven immunosuppression, reshape the immune TME, and elicit robust antitumor immunity. This rationale is supported by the therapeutic success of liposomal muramyl tripeptide phosphatidylethanolamine (L-MTP-PE), a muramyl dipeptide analog that activates macrophages by binding to the intracellular nucleotide-binding oligomerization domain containing protein 2.[36]9,[37]12 Administration of L-MTP-PE as an immunotherapy improved event-free and overall survival in patients with metastatic OS and was associated with the activation of pulmonary macrophages.[38]^9 10 Therapeutic success was also seen in a Phase III trial of patients with newly diagnosed OS.[39]^11 12 Here, we investigated the combination of two innate immune activators, yeast-derived particulate β-glucan and CD40 agonist, to enhance antitumor immunity against OS lung tumor burden. β-glucans are polysaccharides on cell walls of fungi, bacteria, and plants. β-glucans are recognized as pathogen-associated molecular patterns (PAMPs) that activate macrophage via Dectin-1 receptor.[40]^13 Activation of macrophages by β-glucan has been shown to increase phagocytic potential, induce trained immunity, and enhance antitumor immunity.[41]^13 14 Additionally, CD40 is a member of the TNF superfamily and is expressed in many innate immune cells. CD40-CD40L signaling is essential for optimal activation of antigen presenting cells (APCs), such as dendritic cells (DCs) and macrophages, and has been shown to enhance cross-presentation of antigens.[42]^15 16 Furthermore, CD40 agonist Ab (CD40a) administration has been shown to activate APCs and stimulate antigen-specific T cells.[43]^16 Using an experimental metastasis model, we first evaluated the antitumor activity of a yeast-derived particulate β−1,3/1,6 glucan against established OS lung tumors. In the pretreatment settings, administration of β-glucan prior to tumor cell injection significantly inhibited the development of lung tumors and improved overall survival. Treatment of tumor-bearing mice with β-glucan alone significantly improved overall survival and increased the number of activated macrophages and NK cells in the lungs. The combination of CD40a with β-glucan further improved antitumor response and overall survival compared with monotherapy. This therapeutic benefit was associated with increased infiltration of activated macrophages and NK cells in the lungs compared with monotherapy, supporting the immune modulatory potential of this combination therapy. Methods Animal studies Male and female BALB/c and C57BL/6 mice (6–8 weeks old) were purchased from Charles River at NCI Frederick. To assess efficacy against established metastases, BALB/c male and female mice were intravenously injected with 0.5×10^6 K7M3-luc cells. Treatment began according to the schema in [44]figure 1a. Tumor burden was assessed by bioluminescence in vivo imaging systems (IVIS). Mice were injected with D-luciferin (GoldBio 150 mg/kg) intraperitoneally (i.p.), every week. Mice were followed for tumor burden and survival over time. C57BL/6 mice were intravenously injected with 0.5×10^6 F420 cells and monitored for survival outcomes. Mice were randomized post-tumor cell injection before receiving therapy. Mice exhibiting signs of morbidity (loss in body weight >20%, hyperpnea, hunched posture) were euthanized. Figure 1. Pretreatment with β-glucan prevented OS lung metastasis. (a) Experimental schema. (b–e) BALB/c female mice were pretreated with PBS (n=4) or β-glucan (n=4) 7 days prior to 5×10^5 K7M3-luc OS cell injection. (b) Tumor flux over time. (c) Representative bioluminescence imaging image on day 30 post-tumor cell injection. (d) Survival curve analysis of female mice pretreated with PBS and β-glucan followed by injection with K7M3-luc cells. Graphs representative of two independent experiments. (e) Tumor area, number of metastases and representative lung H&E images on day 30. H&E-stained lung sections showing all five lobes from a single representative mouse per group. Graphs representative of two independent experiments. (f) Tumor flux over time in BALB/c male mice pretreated with PBS (n=7) or β-glucan (n=7) 7 days prior to K7M3-luc OS cell injection (5×10^5, intravenous). (g) Survival curve of BALB/c male mice pre-treated with PBS or β-glucan 7 days prior to 5×10^5 K7M3-luc OS cell injection. Graph representative of two independent experiments. (h) Representative flow cytometry plots and frequencies of tumor cells (CD45^−CFSE^+) in the lungs of BALB/c female mice pretreated with PBS or β-glucan 7 days prior to CFSE^+ tumor cell injection. 24 hours post-tumor cell injection lungs were analyzed for tumor cells in the lungs. Graphs representative of two independent experiments. Data shown as mean±SEM. For statistical analysis, two-way analysis of variance with Tukey’s post hoc test (b), Mann Whitney test (e, f, h) and log-rank test (d, g) was used. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. CFSE, Carboxyfluorescein sccinimidyl ester; OS, osteosarcoma; PBS, phosphate-buffered saline. [45]Figure 1 [46]Open in a new tab To evaluate if pretreatment with β-glucan prevents the development of lung metastases, BALB/c male and female mice were first treated with β-glucan (2 mg, i.p.). 7 days after β-glucan treatment, mice were injected (intravenously) with 0.5×10^6 K7M3-luc cells. Bioluminescence IVIS imaging was done every week to assess tumor burden over time and mice were followed for survival. To assess if β-glucan pretreatment inhibits tumor cell engraftment, BALB/c mice were treated with β-glucan (2 mg, i.p.) or phosphate-buffered saline (PBS). 7 days later, mice were injected intravenously with 1×10^6 carboxyfluorescein sccinimidyl ester (CFSE) (5 µM) labeled K7M3-luc cells (intravenously). After 24 hours, mice were euthanized, and lung tissues were harvested and analyzed for the presence of tumor cells by flow cytometry. Preparation of β-glucan and treatment Commercially available β-glucan (Wellmune) derived from Saccharomyces cerevisiae was used in the study. Wellmune is a whole yeast 1,3/1,6 β-glucan that contains 80% β-glucan and has been studied in multiple clinical trials.[47]17,[48]19 Particulate β-glucan was suspended in sterile PBS (4 mg/mL) and then sonicated using a Misonix S-3000 sonicator for 10 cycles of 10 s each with a 1 min interval on ice to ensure uniform dispersion. For treatment, mice received 2 mg of sonicated β-glucan suspension i.p. or 0.5 mg intravenously on day 4 post-tumor cell injection. For oral β-glucan treatment, animals received 2 mg of sonicated β-glucan suspension in PBS, administered 5 days per week (2 days off) for a total of 3 weeks. Control animals received PBS. For CD40a treatment, mice received 50 µg anti-mouse CD40 antibody i.p. (Bio X Cell, FGK4.5), at day 10 post-tumor cell injection. Tissue processing for single-cell suspension Lungs, lungs draining cervical lymph nodes (LDLN) and spleen were harvested from mice. Single-cell suspensions were prepared by mashing through a 70 µm cell strainer. For lung tissues, all lobes except the right caudal lobe were collected for flow analysis. The tissues were finely chopped and enzymatically digested in complete Roswell park memorial institute (RPMI) media supplemented with collagenase-IV (2 mg/mL; Worthington) and DNAse (Sigma) at 37°C for 30 min. The digested tissues were then mashed through a 70 µm cell strainer to obtain single-cell suspension. Cells were centrifuged, aspirated and red blood cell (RBC) lysis (BioLegend) was performed. The cells were quenched with complete media and resuspended in complete media for downstream analysis. Spleen processing was performed as described for lymph nodes, with the additional step of RBC lysis. After RBC lysis, the cells were quenched in complete media, centrifuged, and resuspended in complete media. Statistical analysis All data points presented are mean±SEM. Statistical analysis was determined using Mann-Whitney U test for two-group comparison, or two-way analysis of variance with post hoc Tukey’s test for multiple group comparison. Survival analysis was done by using Kaplan-Meier plots, and statistical significance was determined by using the log-rank test. All statistical analysis was performed by using GraphPad Prism. Multiple comparisons of survival curves were performed in R (V.4.3.2) using the survminer package using log-rank test with Bonferroni-Holm method of correction.[49]^20 P value<0.05 was considered statistically significant. Results Pretreatment with β-glucan induces protective antitumor response and increases overall survival β-glucan stimulated innate immune cells have been shown to undergo epigenetic and metabolic alterations, leading to long-term functional adaptation.[50]^13 21 This results in an increased immune response to an unrelated secondary challenge, termed as “Trained Immunity”, which has been shown to enhance immune surveillance against cancers.[51]22,[52]25 We investigated whether β-glucan pretreatment impairs lung colonization and tumor establishment by OS cells in an experimental metastasis model. BALB/c female mice were treated with PBS or β-glucan (2 mg, i.p.). At day 7 post β-glucan treatment, mice were intravenously injected with K7M3-luc (0.5×10^6) OS cells and followed for tumor growth and survival ([53]figure 1a). β-glucan pretreatment significantly inhibited tumor growth, reduced tumor burden, and prolonged survival ([54]figure 1b–d). At day 35 post-tumor cell injection, β-glucan pretreated mice had significantly less tumor burden as compared with the PBS pretreated mice (p<0.0001, [55]figure 1b and c). All PBS pretreated mice died by day 42. By contrast, all mice pretreated with β-glucan before tumor cell injection survived for >100 days (p=0.0091, [56]figure 1d). Histological analysis of lungs at day 30 post-tumor cell injection revealed that mice pretreated with β-glucan had no evidence of visible lung tumor nodules ([57]figure 1e). Sex has been shown to play a role in immunotherapy success and treatment outcome.[58]^26 We therefore investigated whether the preventive effect by pretreating with β-glucan was also observed in males. BALB/c male mice were pretreated with β-glucan followed by K7M3-luc OS cells injection after 7 days as described above ([59]figure 1a). We observed a similar antitumor effect by pretreating BALB/c male mice with β-glucan as was seen in female mice ([60]figure 1f and g). BALB/c male mice pretreated with β-glucan had significantly reduced tumor burden ([61]figure 1f) and significantly improved overall survival ([62]figure 1g). We next investigated whether β-glucan pretreatment affected the seeding and engraftment of tumor cells into the lungs. PBS-pretreated or β-glucan-pretreated female mice were injected (intravenously) with CFSE-labeled K7M3-luc cells. Lungs were harvested 24 hours later and analyzed for tumor cells. Flow cytometric analysis revealed that lungs from β-glucan-treated mice had significantly fewer CFSE-labeled tumor cells ([63]figure 1h). This suggests that β-glucan pretreatment affected the engraftment of tumor cells into the lungs. Taken together, these data suggest that β-glucan pretreatment increased immune surveillance resulting in reduced tumor cell engraftment, decreased tumor growth and increased overall survival. This preventive effect of β-glucan is sex unbiased with similar antitumor effect in both males and females. β-glucan induces antitumor response against established OS lung metastases BALB/c female mice were intravenously injected with 0.5×10^6 K7M3-luc OS cells. After confirming the presence of OS tumors in the lungs by imaging on day 3, mice were treated i.p. with PBS (Ctrl) or β-glucan (2 mg) ([64]figure 2a). Single-dose treatment of β-glucan significantly inhibited the growth of OS lung tumor and improved the overall survival ([65]figure 2b–e). Mice treated with β-glucan had significantly reduced total flux as a measure of tumor burden (p<0.0001) at day 28 ([66]figure 2b and c). By day 28 in the PBS-treated mice, there was a 430-fold increase in tumor burden (from the reference time of day 3 before treatment) as compared with a 93.3-fold increase in the β-glucan treated mice (p=0.0159, [67]figure 2d). The antitumor effect of β-glucan resulted in a significant increase in overall survival ([68]figure 2e). PBS-treated mice had median survival time (MST) of 33 days, compared with 43 days in the β-glucan treated mice (p=0.0034, [69]figure 2e). We also performed histological analysis of the tumor-bearing lungs from PBS and β-glucan treated mice at day 30 post-tumor cell injection. Histological analysis showed that mice in the β-glucan treated group had a significant reduction in total area of tumor in lungs (p=0.0317) and significantly fewer lung tumor nodules (p=0.0079) as compared with control group ([70]figure 2f). We validated the antitumor therapeutic effect of β-glucan against established lung metastases in males ([71]figure 2g–i). Tumor-bearing BALB/c male mice treated with β-glucan had a significant reduction in tumor burden at day 30 ([72]figure 2g and h). The β-glucan treatment also significantly prolonged survival and increased the MST (p=0.0005, [73]figure 2i). Taken together, these findings suggest that β-glucan had antitumor activity against established OS tumors in the lung resulting in reduced tumor growth and significantly prolonged overall survival. The therapeutic effect is sex unbiased as the effect was observed in both female and male mice. Figure 2. β-glucan treatment inhibits progression of osteosarcoma lung metastasis. (a) Experimental schema. (b–f) K7M3-luc (5×10^5) cells were injected intravenously into BALB/c female mice. Mice were treated 4 days later with PBS (n=5) or β-glucan (n=5). (b) Tumor flux over time. (c) Representative bioluminescence imaging (BLI) showing tumor burden at day 28 post-tumor cell injection. (d) Fold change in BLI on day 28 with respect to day 3 baseline before treatment. (e) Kaplan-Meier survival curve of mice treated with β-glucan or PBS. Graphs representative of three independent experiments. (f) Quantification of tumor area, number of metastases and representative H&E image of lungs on day 30 post-tumor cell injection. H&E-stained lung sections showing all five lobes from a single representative mouse per group. Graphs representative of two independent experiments. (g–i) K7M3-luc injected BALB/c male mice were treated with PBS (n=7) or β-glucan (n=8). (g) Tumor flux over time in BALB/c male mice treated with β-glucan or PBS. (h) Fold change in BLI on day 30 relative to day 3 as baseline before treatment. (i) Survival curve of tumor-bearing BALB/c male mice treated with β-glucan or PBS. Graphs representative of two independent experiments. Data shown as mean±SEM. For statistical analysis, two-way analysis of variance with Tukey’s correction in (b, g), Mann-Whitney test (d, f, h) and log-rank test (e, i) was used. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. PBS, phosphate-buffered saline. [74]Figure 2 [75]Open in a new tab β-glucan treatment activates innate immune cells and induces myelopoiesis We next investigated the immune composition of the lung, LDLNs and spleen from PBS and β-glucan-treated tumor-bearing mice. Flow cytometric analysis on day 10 post-treatment revealed that in comparison to PBS-treated control mice, β-glucan-treated mice had a significant increase in CD11b^+ myeloid cells in the lungs ([76]figure 3a). This was associated with the increased frequency of macrophages (CD11b^+Ly6G^−F4/80^+) and pro-inflammatory M1-like macrophages co-expressing major histocompatibility complex (MHC)-II and CD80, monocytic dendritic cells (mDCs, CD11b^+CD11c^+), and activated mDCs expressing MHC-II and CD80 ([77]figure 3a). Furthermore, macrophages and mDCs in the lungs of β-glucan-treated mice had significantly increased expression of MHC-II and CD80 ([78]figure 3a), which are required for antigen presentation and priming of adaptive immunity, respectively. In parallel, we also found a significant increase in the frequency of CD49b^+NKG2D^+ NK cells and increased expression of NKG2D in NK cells in mice treated with β-glucan ([79]figure 3b). This was also associated with an increased level of cytotoxic NK cell secreting interferon (IFN)-γ (CD49b^+IFN-γ^+) or granzyme B (CD49b^+granzymeB^+). There was also an increase in mean fluorescence intensity (MFI) of IFN-γ and granzyme B in NK cells ([80]figure 3b). Additionally, there was a significant increase in activated myeloid cells (macrophages, mDCs) and NK cells in the LDLNs and spleen ([81]online supplemental figure S1 and S2). These data suggest that β-glucan treatment induced a systemic increase in pro-inflammatory cytotoxic innate immune cells. Figure 3. Effect of β-glucan on lung immune phenotype and bone marrow. (a, b) Flow cytometric analysis of lungs from tumor-bearing mice treated with PBS or β-glucan on 10 days post β-glucan treatment for (a) total myeloid cells (CD11b^+), macrophages (CD11b^+Ly6G^−F4/80^+), M1-like macrophage expressing both MHC-II and CD80 (CD11b^+Ly6G^−F4/80^+MHC-II^+CD80^+), monocytic DC (CD11b^+CD11c^+), activated monocytic DC expressing both MHC-II and CD80 (CD11b^+CD11c^+MHC-II^+CD80^+) and (b) NK cells including activated NK (CD49b^+NKG2D^+), IFN-γ secreting NK cells (CD49b^+IFN-γ^+) and granzyme B secreting NK cells (CD49b^+GranB^+). (c, d) Flow cytometric analysis of BM harvested from tumor-bearing mice treated with PBS or β-glucan 10 days post-treatment. (c) Representative flow cytometry plots and (d) frequencies of LSK (Lin^−Sca1^+c-kit^+), MPP2 (Flt3^−CD48^+CD150^+LSK), MPP3 (Flt3^−CD48^+CD150^−LSK) and MPP4 (FLT3^+CD48^+CD150^−LSK) cell. (e–g) Adoptive transfer of bone marrow cells from PBS or β-glucan treated mice to tumor bearing mice. (e) Experimental schema. (f) Tumor flux, and representative bioluminescence imaging images on day 30 post-tumor cell injection. (g) Survival curve of tumor-bearing recipient mice adoptively transferred with bone marrow cells (10×10^6) from β-glucan-treated or PBS-treated donor mice. Graphs representative of two independent experiments. Data shown as mean±SEM. Mann-Whitney test (a, b, d) and log-rank test (g) was used for statistical significance. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. BM, Bone Marrow; DC, dendritic cell; IFN, interferon; MFI, mean fluorescence intensity; MHC, major histocompatibility complex; NK, natural killer; PBS, phosphate-buffered saline. [82]Figure 3 [83]Open in a new tab To determine which effector immune cells were critical for the β-glucan-induced antitumor response, we performed a series of depletion studies. Macrophages, NK cells, neutrophils or CD4/CD8 T cells were depleted by injecting mice with clodronate, anti-asialoGM1, anti-Ly6G or anti-CD4/CD8 antibodies, respectively, starting 24 hours prior to β-glucan treatment ([84]online supplemental figure S3). Depletion of macrophages or NK cells significantly abrogated the β-glucan-induced protective effect against established OS lung tumors. As expected, mice treated with β-glucan had significantly prolonged survival (p=0.0019). By contrast, the survival of β-glucan-treated mice depleted of macrophages or NK cells was not significantly different from the control mice ([85]online supplemental figure S3a and b). However, the depletion of neutrophils or T cells did not impair the antitumor activity of β-glucan ([86]online supplemental figure S3c and d). These data indicate that the antitumor effect of β-glucan was mediated by macrophages and NK cells, but independent of neutrophils and T cells. Since we observed a significant increase in myeloid cells in the lungs, LDLNs and spleen after β-glucan treatment, and active infection is known to induce emergency myelopoiesis,[87]^27 28 we next determined whether β-glucan treatment mimics an infection by increasing myeloid progenitors in the bone marrow. Tumor-bearing mice were treated with β-glucan or PBS. Bone marrow (BM) cells were harvested 10 days post therapy and analyzed for hematopoietic progenitor cells (LSK: Lin^−Sca-1^+c-kit^+), myeloid progenitor cells (MPP3: Flt3^−CD48^+CD150^−LSK; MPP2: Flt3^−CD48^+CD150^+LSK) and lymphoid progenitor cells (MPP4: FLT3^+CD48^+CD150^−LSK) as previously described.[88]^28 29 β-glucan treatment significantly increased LSK progenitors, and myeloid progenitor MPP3 and MPP2 cells ([89]figure 3c and d). While we observed a slight increase in lymphoid-primed MPP4 lineage cells (FLT3^+CD48^+CD150^−LSK), this was not significant. We also demonstrated a similar effect of β-glucan in inducing myelopoiesis in naïve mice without tumors ([90]online supplemental figure S4). These results are similar to other reports showing β-glucan-induced myelopoiesis as a mechanism for “Trained immunity” against cancer.[91]^21 24 29 We next investigated whether the β-glucan-induced myeloid progenitor cells in the BM contributed to its antitumor effect. BM cells were isolated from β-glucan or PBS-treated mice at 7 days post treatment and adoptively transferred to tumor-bearing irradiated recipients (800 cGY, [92]figure 3e). On day 30 post-tumor cell injection, β-glucan-BM recipients had reduced tumor burden compared with PBS-BM recipients ([93]figure 3f). Furthermore, β-glucan-BM transfer also resulted in a significantly increased overall survival and extended the MST (64 vs 48 days, p=0.0127) ([94]figure 3g). These findings suggest that β-glucan treatment enhances myelopoiesis and specific myeloid progenitor cell populations in the BM and that these BM cells played a role in inhibiting OS lung tumor growth. Taken together, these data indicate that β-glucan treatment increased cytotoxic pro-inflammatory macrophages, mDCs and NK cells in the tumor through expansion of hematopoietic progenitors in bone marrow and that the therapeutic antitumor response is mediated by macrophages and NK cells. Effect of CD40 agonist on the antitumor efficacy of β-glucan Since β-glucan treatment increased lung infiltration with activated macrophages and NK cells ([95]figure 3a and b), we next determined whether an immune agonist that activates innate immune cells further enhanced this antitumor activity. CD40 agonist (CD40a) activates macrophages, DCs, NK cells and is critically important in enhancing cross-presentation. Multiple studies have reported activation of the CD40-CD40L pathway as a molecular adjuvant for cancer immunotherapies.[96]^16 We therefore sought to determine if combining CD40a with β-glucan treatment will result in an enhanced antitumor response. Tumor-bearing mice were treated with β-glucan on day 4 and/or CD40a on day 10 post-tumor cell injection. Tumor burden and survival were quantified ([97]figure 4a). CD40a and β-glucan alone inhibited tumor growth and prolonged survival similarly ([98]figure 4b–e). The tumor growth rate was significantly decreased and survival increased in mice treated with combination therapy compared with monotherapy ([99]figure 4b–e). At 30 days post-tumor cell injection, the average tumor burden in the control group increased by 375-fold. In comparison, mice treated with β-glucan or CD40a alone exhibited a 90.6-fold and 94.8-fold increase in tumor burden, respectively, while the combination treatment group showed a 15.5-fold increase in tumor growth ([100]figure 4d). Survival curve analysis revealed that treatment with β-glucan or CD40a resulted in similar survival rates with a significant increase in MST (52 days) as compared with PBS-treated control cohort (42 days). Combination therapy further improved the overall survival and increased the MST to 59 days ([101]figure 4e). Figure 4. Antitumor effect of β-glucan in combination with CD40a. (a) Experimental schema. (b–e) K7M3-luc-injected BALB/c mice were treated with PBS, β-glucan, CD40a or β-glucan+CD40 a (n=5/group). (b) Tumor flux over time. (c) Representative BLI on day 30. (d) Fold change in BLI on day 30 with respect to day 3 (prior to treatment); and (e) Survival curve. (f–h) F420-injected C57BL/6 mice were treated with PBS, β-glucan, CD40a or β-glucan+CD40 a (n=5/group). (f) Survival curve, (g) representative H&E images; H&E-stained lung sections showing all five lobes from a single representative mouse per group. (h) Tumor area and (i) number of lung metastases on day 30 post-tumor cell injection. Data shown as mean±SEM. For statistical significance, two-way analysis of variance with Tukey’s post hoc test in (b), Mann-Whitney test (d, h, i) and log-rank test with Bonferroni-Holm method of correction (e, f) was used. *p<0.05, **p<0.01, ***p<0.001. BLI, bioluminescence imaging; PBS, phosphate-buffered saline. [102]Figure 4 [103]Open in a new tab We also evaluated the therapeutic effect of combination treatment in a second OS lung experimental metastasis model using F420 OS cells in C57BL/6 mice. F420 cells were injected (intravenously) into C57BL/6 female mice. Mice were then treated with β-glucan and/or CD40a as described above ([104]figure 4a). β-glucan or CD40a treatment significantly improved mouse survival and extended the MST (β-glucan: 50; CD40a: 47 days) as compared with PBS treated control (41 days) ([105]figure 4f). Once again, combination treatment significantly enhanced the overall survival and further extended MST to 60 days ([106]figure 4f). Histological analysis further corroborated the findings. Lungs from β-glucan or CD40a-treated mice had significantly fewer lung tumors and a reduced tumor area compared with the control mice ([107]figure 4g–i). Mice treated with both agents had significantly less tumor area and number of tumor nodules as compared with monotherapy treated mice ([108]figure 4h and i). Taken together, these data demonstrate that combining CD40a with β-glucan increased antitumor activity and improved survival. Effect of combined treatment on the activation of macrophages and NK cells Flow cytometric analysis on tumor-bearing lungs was performed 14 days post-tumor cell injection. The combination therapy significantly increased infiltration of CD11b^+ myeloid cells, macrophages (CD11b^+Ly6G^−F4/80^+), and pro-inflammatory M1-like macrophages (CD11b^+Ly6G^−F4/80^+MHC-II^+CD80^+), as compared with mice treated with single agents ([109]figure 5a), and also significantly increased TNF-α secreting macrophages (CD11b^+F4/80^+TNF-α^+) and expression of TNF-α as quantified by TNF-α MFI compared with monotherapy ([110]figure 5b and c). However, the monocytic DCs (mDCs, CD11b^+CD11c^+) and mDCs co-expressing MHC-II and co-stimulatory CD80 were not significantly increased compared with single agent β-glucan treatment alone ([111]figure 5d). These data show that the synergistic therapeutic effect of β-glucan and CD40a correlated with the increased level of activated proinflammatory macrophages. We also observed that co-treatment significantly altered the NK cell phenotype in the lungs. Compared with single agent-treated mice, the lungs of mice that received both agents had significantly more activated NK cells (CD49b^+NKG2D^+) with increased NKG2D expression ([112]figure 5e), an activating receptor in NK cells that regulates the effector function. Furthermore, the combination treatment increased the frequency of cytotoxic NK cells secreting IFN-γ and increased the IFN-γ MFI ([113]figure 5f). While the level of NK cells secreting granzyme B did not significantly increase, the MFI of granzyme B was increased with the combination therapy over monotherapy ([114]figure 5g). This combined therapy also significantly increased polyfunctional cytotoxic NK cells secreting both IFN-γ and granzyme B ([115]figure 5h). To understand whether NK cells or CD4/8 T cells contribute to antitumor efficacy of the combination therapy, each population was selectively depleted using specific antibodies. Depletion of NK cells but not CD4 or CD8 T cells significantly abrogated the protective effect mediated by the combination therapy. Tumor-bearing mice that received combination therapy but were depleted of NK cells exhibited survival response similar to PBS-treated control group, suggesting that NK cells are essential to mediate the antitumor response ([116]online supplemental Figure S5). Figure 5. Immunological changes in the lung induced by single or combination β-glucan and CD40a therapy. Flow cytometric analysis of immune cell populations in lung tissue from tumor-bearing mice. (a) Total myeloid cells (CD11b^+), macrophages (CD11b^+Ly6G^−F4/80^+), and M1-like macrophages (CD11b^+Ly6G^−F4/80^+MHC-II^+CD80^+). (b) Representative flow plot showing CD11b^+TNF-α^+ cells gated in CD11b^+F4/80^+ cells. CD11b-PerCpCy5.5 was used to gate CD11b myeloid cells; frequency of TNF-α secretory macrophages (CD11b^+F4/80^+TNF-α^+). (c) Histogram and geometric MFI of TNF-α in macrophages; (d) monocytic dendritic cells (CD11b^+CD11c^+) and activated monocytic dendritic cell (CD11b^+CD11c^+MHC-II^+CD80^+). (e) Activated NK cells (CD49b^+NKG2D^+) and NKG2D MFI. (f) IFN-γ secreting NK cells (CD49b^+IFN-γ^+) and IFN-γ MFI, (g) granzyme B secreting NK cells (CD49b^+GranB^+) and granzyme B MFI. (h) NK cells co-secreting IFN-γ and granzyme B, MFI for granzyme B and IFN-γ in CD49b^+IFN-γ^+GranB^+ cells in the lungs. Data shown as mean±SEM. For statistical analysis, Mann-Whitney test (a–h) was used. *p<0.05, **p<0.01. IFN, interferon; MFI, mean fluorescence intensity; MHC, major histocompatibility complex; NK, natural killer; TNF, tumor necrosis factor. [117]Figure 5 [118]Open in a new tab We next evaluated the transcriptomic changes in the lungs induced by treatment using bulk RNA sequencing (seq) on day 14 after tumor cell injection. Principal component analysis of the RNA-seq data revealed distinct clustering of transcriptional profiles across treatments ([119]figure 6a). The combination-treated group clustered closely with the β-glucan-treated group, exhibiting reduced variation compared with the CD40a-treated group. This clustering pattern highlights potential similarities in transcriptional responses between the combination and β-glucan treatments. Analysis of differentially expressed genes revealed that in comparison to PBS, administration of β-glucan, CD40a or β-glucan+CD40 a significantly increased 623, 146, and 839 genes, respectively ([120]figure 6b; [121]Online supplemental Figure S6). We found that β-glucan and β-glucan+CD40 a treatments significantly upregulated genes involved in antigen processing and presentation (H2-Q6, Fcgr4, Ciita, B2m), innate immunity (Prf1, Gzmb, Irf8, Tnf, Xcl1) and leukocyte migration (Cxcl1, Cx3cr1, Ccr3, Cxcl10) as compared with PBS-treated mice ([122]figure 6b; [123]online supplemental figure S6). Gene Ontology (Biological process (BP)) analysis indicated that the combination treatment significantly enhanced the immunogenic pathways related to the immune activation (regulation of innate immune response, positive regulation of response to biotic stimulus, regulation of immune effector process), chemotaxis (leukocyte cell–cell adhesion, leukocyte migration), and cytotoxicity-related pathways (leukocyte mediated cytotoxicity, cell killing) ([124]figure 6c; [125]online supplemental figure S7). Heatmap analysis of genes involved in immunogenic pathways (antigen processing and presentation, myeloid activation, pathogen recognition response signaling, activation of innate immune response and leukocyte migration) demonstrated that the combination of β-glucan and CD40a induced higher gene expression compared with single agent treatment ([126]figure 6d). Together, these data demonstrate that the combined therapy with β-glucan+CD40 a alters the lung immune profile by activating innate immunity, enhancing antigen processing and presentation, and promoting recruitment of pro-inflammatory immune cells into the lungs. Figure 6. Bulk RNA-seq transcriptome analysis of lungs. Bulk RNA-seq analysis was performed at day 14 post-tumor cell injection. (a) Principal component analysis plot of RNA transcripts showing variation between therapeutic treatments. (b) Volcano plot showing differentially expressed genes between combination (β-glucan+CD40a) versus PBS ctrl groups at significance determined by adjusted p<0.05 and log2 foldchange >1 or <−1. Highlighted genes shown as: blue: genes associated with antigen processing and presentation; black: genes associated with activation of innate immunity, green: genes associated with leukocyte migration. (c) Gene Ontology: BP pathway enrichment analysis of differentially expressed genes between combination versus PBS ctrl group. (d) Heatmap displaying gene expression associated with immune processes including antigen processing and presentation, myeloid activation, activation of innate immune response and leukocyte migration. BP, Biological Process; PBS, phosphate-buffered saline; RNA-seq, RNA sequencing. [127]Figure 6 [128]Open in a new tab Intravenous β-glucan but not oral β-glucan elicited an antitumor immune response β-glucan is a PAMPs recognized by innate immune cells. Particulate β-glucans are actively phagocytosed by macrophages to elicit the immune response against tumors.[129]^13 I.p. β-glucans are primarily phagocytosed by peritoneal macrophages to elicit activated immune response. Since i.p. injection is not a standard treatment procedure, we next investigated if a similar antitumor immune response is activated following oral or intravenous administration of β-glucan. While we did not observe the induction of an antitumor effect when β-glucan was administered by oral gavage ([130]online supplemental figure S8), the intravenous administration of β-glucan alone at a reduced dose (0.5 mg), or in combination with CD40a, induced an antitumor immune response similar to that achieved with i.p. β-glucan and CD40a ([131]figure 7). Figure 7. Therapeutic effect of β-glucan administered intravenously in combination with CD40a. (a) Experimental Schema. (b) Representative bioluminescence imaging and fold change of tumor flux at day 30 post-tumor cell injection with respect to day 3 (before treatment). (c) Survival curve of mice treated with PBS, β-glucan (intravenously), CD40a or combination (β-glucan (intravenously) + CD40 a). Data shown as mean±SEM. For statistical analysis, Mann-Whitney test (b) and log-rank test with Bonferroni-Holm method of correction (c) was used. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. IVIS, In vivo imaging systems; PBS, phosphate-buffered saline. [132]Figure 7 [133]Open in a new tab Discussion Activation of innate immunity shapes the antitumor response.[134]^30 31 Pattern recognition receptors on innate immune cells (DCs, macrophages, NK cells and neutrophils) detect PAMPs, triggering innate immunity and inducing non-specific immune responses against infection and tumors.[135]^32 33 Innate immune cells also exert direct antitumor effects, including macrophage-mediated cytotoxicity and phagocytosis, and NK cell-mediated cytotoxicity via perforin and granzyme.[136]^30 APCs, DCs and macrophages, form a crucial link between innate and adaptive immunity. Their activation and maturation significantly enhance antigen processing and presentation capabilities, activating tumor-specific T-cell responses.[137]^34 β-glucan has emerged as a potent immunomodulator with antitumor properties. The binding of β-glucan to the Dectin-1 receptor expressed on monocytes, macrophages, DCs, and neutrophils triggers their activation and induces ROS generation, cytokine secretion and microbial killing.[138]^13 Using an experimental metastasis model, we demonstrated the antitumor effect of yeast-derived particulate β-glucan. In the pretreatment settings, the administration of β-glucan prior to tumor cell injection prevented engraftment of OS tumor cells in the lung, resulting in significantly fewer lung metastases and a reduction in tumor size. This effect also led to a significantly prolonged overall survival. β-glucan therapy was also effective against established OS lung tumors. When β-glucan was administered after tumor cell injection to mice with metastases present in the lung, there was once again a reduction in tumor size and the number of lung tumor burden, as well as a significantly prolonged overall survival. These antitumor effects of β-glucan were sex unbiased, as both preventive and therapeutic efficacy were observed in both female and male mice. Our results demonstrated that the antitumor effect correlated with alteration of the lung immune cell microenvironment. The therapeutic effect was accompanied by an increase in activated macrophages, monocytic DCs and NK cells in the tumors of β-glucan-treated mice. Immune depletion studies showed that both macrophages and NK cells were critical to the β-glucan induced antitumor effect. The selective depletion of either macrophages or NK cells abrogated β-glucan’s therapeutic effect. These results are consistent with previous studies demonstrating the importance of innate immune cells in β-glucan-induced antitumor efficacy.[139]^21 35 While previous studies suggest macrophages as the primary effector cells in β-glucan-mediated immunity, our data reveal a critical role for NK cells as well, highlighting the importance of both macrophages and NK cells in mediating its antitumor effects. Since β-glucan treatment induced a significant increase in myeloid cells in the lungs, we hypothesized that these myeloid cells were coming from the BM and investigated its effect on BM cell composition. We observed an increase in myeloid progenitors in the BM of β-glucan-treated mice compared with PBS-treated controls, suggesting that β-glucan treatment stimulated BM myelopoiesis. Furthermore, the adoptive transfer of BM cells from β-glucan-treated tumor-free mice into tumor-bearing recipients resulted in reduced tumor growth and improved overall survival. This effect was not seen when BM cells from PBS-treated mice were transferred into the tumor-bearing mice. Taken together, these data suggest that the β-glucan-induced myeloid progenitors in the BM migrated to the lungs, differentiating into the pro-inflammatory antitumor innate immune cells that contributed to the β-glucan-induced tumor suppression. The mechanism that controls the programming and recruitment process of myeloid precursors from the bone marrow into the OS lung tumor is unclear at this time but may involve specific integrins such as α[5]β[1], and αvβ[3], which have been shown to be involved in breast cancer models.[140]^36 Since β-glucan treatment increased the number of activated macrophages, monocytic DCs and NK cells in the lungs, and its antitumor activity was dependent on macrophages and NK cells, we hypothesized that combining β-glucan with an activator of the innate immune pathway, such as the CD40-CD40L pathway, would further enhance its antitumor immunity and therapeutic efficacy. CD40, a member of the TNF receptor superfamily, is expressed on activated APCs and plays a pivotal role in linking innate and adaptive immunity. Specifically, crosslinking of CD40L on activated CD4^+ T cells to CD40 on DCs has been shown to be a critical step in “licensing” DCs for efficient antigen presentation and subsequent activation of antigen-specific CD8^+ T cells.[141]^16 37 CD40 agonist antibody has been shown to activate tumor-specific immune responses and synergize with immune checkpoint blockade, chemotherapy, and radiotherapy.[142]^16 Using two different syngeneic mouse models, we showed that combining β-glucan and CD40a resulted in an enhanced antitumor response and superior overall survival compared with monotherapy. The combination therapy of β-glucan+CD40 a significantly altered the lung immune phenotype. Flow cytometric analysis revealed that the lungs from mice treated with both agents had a significantly higher frequency of CD11b^+ myeloid cells and activated M1-like macrophages compared with monotherapy. Moreover, we also observed an increase in TNF-α secreting macrophages and higher TNF-α MFI in the lungs from mice treated with the combination therapy. NK cells in the β-glucan+CD40 a group were also more activated, showing higher levels of IFN-γ and granzyme B. There was also an increase in the population of polyfunctional NK cells secreting both cytokines. Bulk RNA-seq analysis of lungs from treated mice also suggested that combination therapy with β-glucan+CD40 a induced a superior pro-inflammatory response in the lungs. This was evidenced by increased expression of genes involved in immunogenic pathways including antigen processing and presentation, myeloid activation, innate immune activation, and leukocyte migration in the combination group compared with monotherapy. Overall, combination therapy induced higher gene expression compared with single-agent treatment. Our findings have important clinical implications. There have been no significant advancements in treating patients with OS with lung metastases either in the newly diagnosed or relapsed setting. Immunotherapeutic strategies have primarily focused on activating T-cell responses through immune checkpoint inhibitors, but these have been largely ineffective due to the immune suppressive microenvironment in the lung metastases.[143]^1 38 Here we show that innate immune cells can be targeted using β-glucan to generate an antitumor immune response, and that combining β-glucan and another innate immune activator, CD40 agonist, enhanced therapeutic activity against established OS lung tumors and altered the tumor immune microenvironment. In addition, we showed that β-glucan prevented metastatic spread to the lungs by inhibiting tumor cell engraftment. These findings suggest that in addition to treating relapsed patients, β-glucan may have a role for patients after preoperative chemotherapy and surgery during the surgery recovery/healing period when no chemotherapy is administered. This is a vulnerable time for the patient as there is a correlation between the length of time before resumption of chemotherapy and survival.[144]^39 Initiating β-glucan therapy may reduce the relapse rate, which is now 30%–35%. Targeting innate immune cells, particularly the myeloid population, offers an advantage since their activation enhances antitumor responses and downregulates the immune suppression that limits T-cell-based immunotherapies.[145]^40 In our study, administration of β-glucan prior to tumor cells led to durable protection in a subset of mice, with survival extending beyond 100 days. This may indicate that β-glucan therapy is most effective in the setting of minimal residual disease similar to what was shown with L-MTP-PE.[146]^9 10 Survival results were not as good in the therapeutic setting as we only administered a single dose of β-glucan. Increasing the number of β-glucan treatments for a longer time period may improve efficacy and result in a higher percentage of durable responses. We also anticipate that combining this approach with other T-cell-based therapies may improve outcomes. The safety and tolerability of the CD40 agonist has already been established in pediatric patients.[147]^41 No information is yet available, however, regarding the tolerability of β-glucan. In summary, our investigations indicate that β-glucan and the combination of β-glucan plus CD40 agonist has potential in the treatment of patients with OS with relapsed, refractory or metastatic disease. Supplementary material online supplemental file 1 [148]jitc-13-10-s001.docx^ (4.4MB, docx) DOI: 10.1136/jitc-2025-012510 Footnotes Funding: This study was supported by a research grant awarded from Enzo & Me PCF to ESK. PS is the recipient of the MD Anderson Odyssey Fellowship at The UT MD Anderson Cancer Center. The funders had no role in study design, data collection, analysis and the interpretation; and in the decision to submit the paper for publication. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: All animal procedures and experiments were performed in compliance with the Animal Care and Use Committee of the UT MD Anderson Cancer Center (000000896RN03). Data availability statement Data are available upon reasonable request. References