CLEVER-1 blockade reprograms TAMs to overcome anti-PD-1 resistance in gastric cancer

Abstract

Common lymphatic endothelial and vascular endothelial receptor-1 (CLEVER-1) is a multifunctional scavenger receptor expressed on tumor-associated macrophages (TAMs). In a recent study published in the Journal for ImmunoTherapy of Cancer, Yu et al reported that CLEVER-1⁺ TAMs accumulate in advanced gastric cancer (GC), associate with poor prognosis, and contribute to resistance to chemoimmunotherapy. CLEVER-1 blockade using bexmarilimab reprogrammed TAMs toward a pro-inflammatory phenotype by suppressing peroxisome proliferator-activated receptor gamma (PPARγ)-driven lipid metabolism and enhancing antigen presentation and inflammatory cytokine secretion. CLEVER-1 blockade also synergized with anti-programmed cell death protein 1 (PD-1) therapy in ex vivo GC models, particularly in tumors enriched with CLEVER-1⁺ TAM. These findings identify CLEVER-1⁺ TAMs as both biomarker and functional mediator of anti-PD-1 therapy resistance, providing a rationale for combining bexmarilimab with immune checkpoint blockade in GC. In this commentary, we discuss the mechanistic significance, translational potential, and clinical prospects of CLEVER-1 blockade to overcome immunotherapy resistance in GC.

Background and significance of the study

Immune checkpoint blockade (ICB) therapies, including anti-programmed cell death protein 1 (PD-1)/programmed cell death ligand 1 (PD-L1) and cytotoxic T lymphocyte-associated protein 4 (CTLA-4) monoclonal antibodies (mAbs), have revolutionized the treatment landscape of many cancers. However, their efficacy remains limited in many solid tumors, including gastric cancer (GC), where only a subset of patients benefits. A major barrier to effective ICB in GC is the presence of immunosuppressive tumor microenvironment (TME), in which tumor-associated macrophages (TAMs) play a pivotal role.¹

In the recent study by Yu et al, common lymphatic endothelial and vascular endothelial receptor-1 (CLEVER-1), a scavenger receptor encoded by STAB1, was identified as a defining marker of a TAM subset enriched in advanced GC.² Using RNA sequencing, multiplex flow cytometry, and multiplex immunofluorescence, the authors characterized CLEVER-1⁺ TAMs at the transcriptomic, phenotypic and spatial levels. These TAMs co-express both M1 and M2 markers but functionally displayed an M2-like phenotype, characterized by high IL-10 and ARG1 expression. This subset was associated with poor prognosis and reduced responsiveness to chemoimmunotherapy. Notably, in some patients who initially responded to treatment, CLEVER-1⁺ TAMs became further enriched after therapy, suggesting their involvement in adaptive resistance.²

Mechanistically, CLEVER-1 expression was associated with a peroxisome proliferator-activated receptor gamma (PPARγ)-dependent lipid metabolism program. Treatment with bexmarilimab reduced the expression of PPARG, IL-10, and ARG1, while enhancing antigen presentation molecules (MHC-II, CD80, CD86), proinflammatory chemokines (CXCL9/10) and cytokines (TNF, IL1B, IFNG). These changes restored CD8⁺ T-cell activation and reversed TAM-mediated suppression in vitro. Furthermore, bexmarilimab synergized with nivolumab in ex vivo GC tissue blocks, increasing cytotoxic T-cell activity and tumor cell apoptosis.² Together, these findings provide proof of concept for a druggable immune-metabolic axis centered on CLEVER-1⁺ TAMs that contributes to ICB resistance in GC.

Mechanistic insights and unanswered questions

Although the study provides strong evidence supporting CLEVER-1 as an immunoregulatory checkpoint in TAMs, several mechanistic and translational questions remain unresolved. The factors driving CLEVER-1⁺ TAM enrichment and polarization are not fully understood. It remains unclear whether these macrophages originate from a distinct lineage or represent a reprogrammed population shaped by tumor-derived or therapy-induced cues. The TME in GC, often marked by hypoxia, acidic stress, and stromal fibrosis, together with soluble mediators such as cytokines, metabolites, and extracellular vesicles, may collectively shape TAM differentiation and activity.³ During chemoimmunotherapy, tissue damage and inflammatory signals released from dying tumor cells may further induce CLEVER-1 expression as a compensatory or adaptive response.⁴ Understanding the timing and triggers of CLEVER-1 induction will be essential for defining the plasticity and stability of these macrophages.

The mechanisms through which CLEVER-1⁺ TAMs suppress antitumor immunity also require further clarification. Both the present and prior studies suggest that these TAMs act through the secretion or release of immunomodulatory factors such as interleukin-10 and transforming growth factor beta (TGF-β). A recent study has indicated that CLEVER-1 itself can be shed as a soluble form (sCLEVER-1), which circulates in plasma and interferes with T-cell activation. Through both secreted cytokines or sCLEVER-1, these TAMs help establish an immunosuppressive TME that restrains effective antitumor responses at local and systemic levels.⁵

CLEVER-1 is part of a large family of scavenger receptors involved in phagocytosis and immune regulation, many of which perform overlapping functions. Examining CLEVER-1 in isolation may therefore overlook the broader complexity of the myeloid regulatory network. Other receptors such as TREM2, TIM-4, MerTK, CX3CR1, CD206 and APOE have been implicated in shaping immunosuppressive TAM phenotypes across various cancer types.⁶ Comparative profiling of CLEVER-1 and these molecules in terms of expression patterns, transcriptional control and downstream pathways could help determine whether CLEVER-1 defines a distinct TAM subset or acts in concert with a broader scavenger receptor network. Such analyses could support refined macrophage taxonomy that moves beyond the traditional M1/M2 framework toward one based on phagocytic and immune-metabolic characteristics.

Within the broader context of TAM-targeted therapies, CLEVER-1 blockade represents a distinct approach compared with existing strategies such as CSF1/CSF1R inhibition and CD47/SIRPα blockade. CSF1R inhibitors primarily act by depleting or preventing the recruitment of macrophages, whereas CD47/SIRPα blockade enhances tumor cells phagocytosis.⁷ In contrast, CLEVER-1 blockade reprograms macrophages toward a pro-inflammatory and antigen-presenting phenotype without eliminating them, thus maintaining their phagocytic capacity while enhancing T-cell activation. This mechanism suggests that CLEVER-1 blockade could enhance rather than compete with other macrophage-targeting therapies. Exploring potential combinations of CLEVER-1 blockade with CSF1R or CD47/SIRPα blockade may help overcome compensatory suppressive pathways and strengthen the durability of immune responses.

The prevalence and function of CLEVER-1⁺ TAMs vary across tumor types and disease stages. While CLEVER-1 expression correlates with immunotherapy resistance and poor prognosis in GC, studies in early-stage colorectal and breast cancers have reported the opposite trend, where higher CLEVER-1⁺ TAMs density associates with better clinical outcomes.^{8 9} These observations suggest that CLEVER-1⁺ TAMs may play context-dependent roles: contributing to inflammation resolution and tissue repair in early disease but promoting immune suppression as tumors progress. Thus, the functional spectrum of CLEVER-1⁺ TAMs likely depends on the local cytokine milieu and immune contexture, shifting from a regulatory phenotype in premalignant settings to a suppressive one in established tumors.

It also remains uncertain whether CLEVER-1⁺ TAMs are consistently distributed across all molecular subtypes of GC. Microsatellite instability-high and Epstein-Barr virus-positive tumors generally exhibit a more inflamed TMEs, while genomically stable and chromosomal instability subtypes tend to be more immune-excluded.¹⁰ Understanding how CLEVER-1 expression varies across these subtypes could clarify its prognostic and predictive roles. Similarly, cross-tumor mapping of CLEVER-1 could clarify whether its role is confined to specific tissue types or reflects a broader hallmark of myeloid-mediated immune resistance.¹¹ Such comparative analyses will be crucial to define whether CLEVER-1 functions as a pan-cancer therapeutic target or a context-specific biomarker.

Methodological considerations and directions for improvement

The ex vivo co-culture system used in this study, which relies on single-cell suspensions, offers several practical advantages. It allows straightforward manipulation of cell populations, direct evaluation of drug effects, and selective depletion of specific subsets such as CD14⁺ myeloid cells through magnetic bead sorting. However, this approach disrupts the native spatial structure and vascular system, limiting its ability to recapitulate the spatial complexity of the TME. As a complementary method, the use of patient-derived tumor fragments, which preserve the three-dimensional structure and immune-stromal interactions of intact tumor tissue, may offer additional translational relevance in future studies.¹²

Another limitation of the study is the absence of immune-competent animal models, which constrains in vivo assessment of CLEVER-1 blockade. Employing chemically induced or syngeneic tumor models would help clarify how CLEVER-1 blockade shapes macrophage behavior and influences T-cell responses within an intact immune system. In addition, humanized mice bearing patient-derived xenografts or organoids could better mimic the therapeutic performance of bexmarilimab in combination with ICBs under a human immune context.¹³

In addition, the clinical cohort analyzed by Yu et al was drawn from a single institution in East China. While this design ensures consistency in clinical management and sample processing, it may not fully represent the full diversity of GC across different populations. Differences in genetic background, environmental exposure, dietary habits, and microbiome composition could influence CLEVER-1 expression and the immune landscape. Broader, multicenter studies across ethnically and geographically diverse populations will be critical to validate these findings and assess their generalizability.¹⁴

Clinical implications, safety concerns, and future directions

CLEVER-1 mAb bexmarilimab is currently being evaluated in several early-phase clinical trials (eg, NCT03733990 and NCT05428969).^{15 16} Most of these studies focus on solid tumors and assess bexmarilimab as monotherapy in advanced or neoadjuvant settings. To date, combination strategies with chemotherapy, radiotherapy, or ICBs have not yet been widely explored. The only registered trial testing bexmarilimab in combination with ICB (NCT05171062, non-small cell lung cancer) has been terminated. Given the growing body of preclinical evidence supporting the synergy between CLEVER-1 blockade and ICBs, expanding clinical research in this direction appears highly promising.¹⁷ Future investigations should also consider evaluating CLEVER-1 blockade in neoadjuvant or maintenance settings and testing its efficacy in biomarker-enriched subgroups to determine whether it can convert immunotherapy-resistant tumors into responders. Comprehensive molecular and immune profiling through pretreatment biopsies, high-throughput sequencing, or multiplex immunohistochemistry may help identify predictive biomarkers such as CLEVER-1⁺ TAM abundance or specific TME signatures. These approaches could enable better patient stratification and facilitate the rational design of combination trials, thereby accelerating the clinical translation of CLEVER-1 blockade.

While the therapeutic rationale for CLEVER-1 blockade is strong, its physiological role in immune homeostasis calls for careful consideration. CLEVER-1 is constitutively expressed on lymphatic and sinusoidal endothelial cells and facilitates the clearance of apoptotic cells and modified self-antigens.¹⁸ Findings from knockout mice have shown that loss of CLEVER-1 can alter lymphocyte trafficking and induce mild tissue inflammation, suggesting a key role in maintaining peripheral immune tolerance.¹⁹ Therefore, long-term or systemic blockade of CLEVER-1 may risk immune-related adverse events, particularly in older patients or those predisposed to autoimmunity. Such risks underscore the need of incorporating ethical safeguards into ongoing and future clinical trials. Patient selection criteria should account for autoimmune history, and close monitoring for immune-related adverse events is warranted. Trial protocols should include longitudinal immune profiling and safety endpoints to evaluate off-target inflammatory effects.

Building on the current evidence, several directions should be prioritized to strengthen clinical relevance and advance future translation. Large multicenter studies are needed to validate CLEVER-1 expression and function across diverse patient populations, molecular subtypes, and ethnic backgrounds. Expanding these investigations to other CLEVER-1-rich malignancies, such as pancreatic, biliary, and ovarian cancers, may further define its value as a predictive biomarker and therapeutic target. Moving forward, well-designed combination trials that integrate CLEVER-1 blockade with ICBs or other TAM-directed therapies should be prioritized. Equally important will be the incorporation of biomarker-based patient selection, molecular profiling, and safety monitoring into trial design. Together, these efforts will provide the clinical and translational foundation to establish CLEVER-1 as both a viable therapeutic target and a clinically meaningful biomarker in GC and beyond.

Conclusion

In conclusion, the study by Yu et al presents compelling mechanistic and translational evidence that CLEVER-1⁺ TAMs promote immune resistance in GC. Their blockade remodels the TME and enhances the efficacy of anti-PD-1 therapy. These findings support continued clinical development of CLEVER-1 blockade therapies and underscore the central role of macrophage plasticity in modulating immunotherapy response. Ongoing research into myeloid plasticity and metabolic regulation may establish CLEVER-1 as a promising therapeutic target in GC and other immune-excluded tumors.