Next-generation CD40 agonists for cancer immunotherapy

ABSTRACT

Introduction

There is a need for new therapies that can enhance response rates and broaden the number of cancer indications where immunotherapies provide clinical benefit. CD40 targeting therapies provide an opportunity to meet this need by promoting priming of tumor-specific T cells and reverting the suppressive tumor microenvironment. This is supported by emerging clinical evidence demonstrating the benefits of immunotherapy with CD40 antibodies in combination with standard of care chemotherapy.

Areas covered

This review is focused on the coming wave of next-generation CD40 agonists aiming to improve efficacy and safety, using new approaches and formats beyond monospecific antibodies. Further, the current understanding of the role of different CD40 expressing immune cell populations in the tumor microenvironment is reviewed.

Expert opinion

There are multiple promising next-generation approaches beyond monospecific antibodies targeting CD40 in immuno-oncology. Enhancing efficacy is the most important driver for this development, and approaches that maximize the ability of CD40 to both remodel the tumor microenvironment and boost the anti-tumor T cell response provide great opportunities to benefit cancer patients. Enhanced understanding of the role of different CD40 expressing immune cells in the tumor microenvironment may facilitate more efficient clinical development of these compounds.

KEYWORDS:

• Antibodies
• antigen-presenting cells
• bispecific antibodies
• CD40
• immuno-oncology
• tumor-associated antigens

1. Introduction

CD40-targeting therapies provide an opportunity to kickstart the cancer-immunity cycle by promoting priming of tumor-specific T cells [1,2], and reverting the suppressive tumor microenvironment (TME) by activation of macrophages and other CD40-expressing immune cell populations [3,4]. Clinical evidence supporting the benefits of immunotherapy with agonistic CD40 antibodies is emerging, in particular as combination with chemotherapy [1,5,6]. By targeting antigen-presenting cells (APCs), CD40 agonists have the potential to synergize both with immune checkpoint inhibitors (ICI) and chemotherapy. Further, CD40 agonists synergize with therapeutic cancer vaccines [1,5].

CD40 is a 48 kDa transmembrane cell surface glycoprotein belonging to the tumor necrosis factor receptor (TNFR) superfamily [7,8]. It is expressed in diverse cell types and can be detected on APCs, including dendritic cells (DCs), macrophages, B cells, and several other cell types such as neutrophils, endothelial cells, smooth muscle cells, fibroblasts, and epithelial cells [7–10]. CD40 is also present on the membranes of a wide range of malignant cells, including B cell malignancies and different carcinomas [8,11]. The CD40 ligand (CD40L) is a transmembrane protein that is expressed primarily by activated T cells, but also by B cells, platelets, mast cells, macrophages, basophils, and natural killer (NK) cells [8,10]. Binding of CD40 to CD40L activates an intracellular signal transduction pathway that involves a series of adapter molecules known as TNFR activation factors (TRAF). Activation of these signal transduction pathways requires formation of superclusters of CD40 [9,12]. This results in intracellular assembly of multiple TRAFs, which in turn leads to the activation of downstream transcription factors, including NF.κB, and subsequently to immune activation [8,13] (Figure 1).

Figure 1. Signaling induced by CD40 bispecific antibodies binding tumor-associated antigens. Upon binding to both of its targets, bispecific antibodies promote CD40 supercluster formation, necessary for downstream signaling. The superclusters of CD40 facilitate the recruitment TNF receptor-associated factors (TRAFs, including TRAF1, TRAF2, TRAF3, TRAF5 and TRAF6). The downstream signaling pathways includes the JAK/STAT pathway, canonical or non-canonical NF-κB pathways or MAPK pathway resulting in immune cell activation or tumor cell apoptosis, if occurring within tumor cells. TAA: tumor-associated antigen; APC: antigen-presenting cell. Created with BioRender.com.

The most studied target cells for CD40 agonists in immuno-oncology are DCs and macrophages. Activation of DCs has the potential to induce priming of tumor-specific T cells and boost the number and fitness of T cells capable of killing tumor cells. Further, CD40 agonists have been found to redirect tumor-infiltrating macrophages from the immunosuppressive M2 phenotype to the pro-inflammatory M1 phenotype [3,4] and to induce activation of myeloid cells in patient-derived tumor samples from both pancreatic cancer and esophageal/gastroesophageal junction cancer [14,15].

The ability of CD40 targeting therapies to stimulate different CD40 expressing immune cells in the TME and other compartments depends on the dose and biodistribution of the CD40 targeting agent. CD40 targeting therapies are in general rapidly eliminated from circulation due to target-mediated drug disposition and short half-life [16–19]. This reduces the biodistribution into lymph nodes and other tissues including the TME. Dose levels that achieve relevant saturation of the CD40 receptor also in the TME may be required to generate efficient CD40 mediated anti-tumor immunity [20], and this needs be considered when designing new CD40 targeting agonistic therapies.

While the first-generation CD40 agonist had tolerability issues, there are currently second-generation monospecific CD40 antibodies in phase 2 development with promising preliminary results [6,14] (Table 1). This review focuses on the coming wave of next-generation CD40 agonists using new modalities beyond monospecific antibodies, in particular bispecific antibodies, and the functional consequences of CD40 stimulation of different immune cell populations in the TME.

Table 1. Monospecific CD40 agonists evaluated in phase 2 studies.

CD40 antibody	IgG subclass	Fc modification	FcγR-conditional agonist	References
Mitazalimab	IgG1	None, wild type IgG1	Yes	[6,18,19,21–23]
Sotigalimab	IgG1	S267E mutation; increased FcγRII binding, reduced FcγRIIIa binding	Yes	[15,24–28]
Giloralimab	IgG1	V273Y mutation; increased FcγRII binding, reduced FcγRIIIa binding	Yes	[29]
Cifurtilimab	IgG1	Afucosylated; increased FcγRIIIa binding	Yes	[30]
CDX-1140	IgG2	None, wild type IgG2	No	[31,32]
Selicrelumab	IgG2	None, wild type IgG2	No	[16,33–37]

2. Next-generation CD40 agonistic therapies for cancer immunotherapy

Novel therapeutic modalities, such as bispecific antibody formats, provide additional opportunities to utilize CD40-targeting to enhance anti-tumor immune responses and direct the resulting immune activation to the TME. Several bispecific CD40-targeting therapies are under development, binding CD40 and a second target. In addition, there are other modalities that use CD40 targeting, such as DC-targeted vaccines and gene therapies. The main driver for this development should be to enhance efficacy while maintaining a good tolerability profile.

2.1. CD40 bispecific antibodies

The functional differences of CD40 bispecific antibodies depend primarily on the choice of CD40 binder and the characteristics of the second target. CD40 therapies can be designed to be active specifically in the TME by selecting an appropriate conditional CD40 agonistic binder [5]. Conditional CD40 agonists only activate CD40 expressing cells while simultaneously binding to a second target, where the second target is required to cluster multiple CD40 receptors and subsequently induce immune activation (Figure 1). The functional effects of conditional CD40 agonistic bispecific antibodies depend on the design and choice of the second binder. Different CD40 bispecific antibodies in development can be categorized based on the nature of the second binding partner as being either tumor associated antigens (TAA)-targeting, stroma-targeting or T cell-targeting CD40 bispecific antibodies (Figure 2 and Table 2).

Figure 2. Three different classes of next-generation CD40 targeting therapies based on the nature of their second target. CD40-bispecific antibodies targeting highly expressed tumor associated antigens (TAA), i.e. Neo-X-Prime bispecific antibodies, results in tumor-direct CD40 activation and enhanced uptake of tumor material hence inducing effective neoantigen cross-priming of tumor specific T cells. Stroma-targeting bispecific CD40 antibodies facilitate tumor-directed activation of CD40. Lastly, T cell-targeting CD40 bispecific antibodies allow for re-activation of tumor specific T cells. DC: dendritic cell; TAM: tumor-associated macrophage; TAA: tumor-associated antigen; TME: tumor microenvironment. Created with BioRender.com.

Table 2. CD40 bispecific antibodies in development divided by subclass. Silent Fc, denotes formats containing an IgG derived Fc containing mutations to reduce FcγR interactions.

CD40 bispecific antibodies class	Drug names	Drug status	Format	Second target	Ref
CD40 × TAA	ATOR-4066	Preclinical	RUBY, tetravalent Fab based, silent IgG1 Fc [38]	CEACAM5	[39]
	Roche bivalent bispecific	Preclinical	CrossMab, silent IgG Fc [40]	CEACAM5	[41]
	KK2269	Clinically active	REGULENT tetravalent Fab-based, common light chain, silent IgG4 Fc	EpCAM	[42,43]
	5T4-CD40	Preclinical	Not determined	5T4	[44]
	Tianjin Medical University General Hospital CD40×HER2 Bs	Preclinical	Tetravalent single chain diabody, no Fc	HER-2	[45,46]
	SL-279252	Clinically Active	ARC- format, hexavalent SIRPa-Fc-CD40L molecule, silent IgG1 Fc	CD47	[47]
	HBM9027	Preclinical	HBICE VH single-domain, silent Fc	PD-L1	[48]
CD40 × Stroma	MP0317	Clinically active	DARPin, tetravalent containing two CD40 binders, one FAP binder and one albumin binding domain	FAP	[49]
	SHR-7367	Clinically active	Hexavalent appended IgG with 4 single domain VHH, silent Fc	FAP	[50]
	RG6189	Discontinued	CrossMab, trivalent with two CD40 binding doimains and one FAP binding domain, silent IgG Fc [40]	FAP	[51,52]
	Affibody Medical AB PDGFRB x CD40	Preclinical	Affimab, tetravalent, silent Fc	PDGFRB	[53]
CD40 × T cell target	YH008/NWY001	Clinically active	Morrison format [54], tetravalent, silent IgG1 Fc	PD-1	[55]
	Biotheus PD-1×CD40 Bispecific antibody	Preclinical	Morrison format [54], tetravalent, silent IgG1 Fc [54]	PD-1	[55,56]
	Tecaginlimab	Clinically active	Duobody, bivalent, silent IgG1 Fc [54]	4-1BB	[57]

Bispecific antibodies targeting CD40 and TAA can enhance efficacy and allow for safe dosing at relevant dose levels by conditionally activating CD40-expressing immune cells in the TME. However, the choice of TAA is key to drive enhanced efficacy. The first CD40 bispecific antibody evaluated in a clinical setting, CD40 × mesothelin bispecific antibody (ABBV-428), displayed a good tolerability profile, but only modest clinical activity [58,59]. This is likely attributed to the low receptor density of mesothelin on the surface of tumor cells [60,61], which is not sufficient to drive an effective anti-tumor response.

One approach that addresses this issue is called Neo-X-Prime®, a bispecific CD40 concept focused on enhancing the anti-tumor activity, while maintaining a good tolerability profile [39]. A critical part of this concept is to select TAAs expressed at high densities, such as CEACAM5 and EpCAM [39]. The aim of this approach is to i) induce strong TAA-conditional CD40 mediated activation of DCs and macrophages in the TME and, ii) increase the uptake of TAA-expressing tumor-derived debris/exosomes/extracellular vesicles, carrying tumor neoantigens, by APCs to promote efficient cross-priming of neoantigen specific T cells. Increased uptake of tumor material and efficient cross-priming of T cells induced by Neo-X-Prime bispecific antibodies have recently been demonstrated by us and others [39,41]. The anti-tumor effect in vivo of a Neo-X-Prime bispecific antibody was shown to be superior to a combination of CD40 and TAA-monotargeting compounds [39].

Bispecific therapies targeting CD40 and CD47 represent another interesting approach to tumor targeted CD40 agonists. CD47, also known as integrin-associated protein (IAP), is a ubiquitously expressed transmembrane receptor overexpressed on some tumor cells [62]. A CD40 × CD47 targeting drug candidate (SL-172154) consisting of a SIRPα-Fc-CD40L fusion protein is currently being developed [47]. In addition to activating CD40-expressing cells, this drug candidate also blocks the CD47/SIRPα-axis and overcomes the ‘don’t-eat-me’ signal induced by CD47 binding to SIRPα. This molecule has been shown to induce superior anti-tumor effects in mouse models compared to the combination of the respective monospecific reagents. The SL-172154 differs from most bispecific antibodies and comprises hexamers of CD40L making it non-conditionally active, i.e. also inducing CD40 agonism in the absence of CD47 binding. This drug candidate has been shown to be safe and tolerable at doses up to 10 mg/kg, and to induce pharmacodynamic responses consistent with CD40 stimulation in cancer patients [63].

PD-L1 targeting CD40 bispecific antibodies can be considered a separate subclass of CD40 × TAA bispecific antibodies, as the primary advantage is related to the function as a PD-1/PD-L1 inhibitor. PD-L1 as a pure tumor target has some limitations as it is expressed on several immune cell populations in addition to tumor cells, including Kupfer cells in the liver [64]. While PD-1/PD-L1 blockade has proven to be very successful, and the combination of CD40 stimulation and PD-1 blockade has a strong preclinical rationale, there are some inherent challenges with this approach based on the short half-life mediated by CD40-binding [5]. This is not an issue for the CD40 agonistic effects, as agonists generally don’t require receptor occupancy over time. In contrast, full receptor occupancy over time of PD-L1 is critical and may not be feasible without dosing very frequently. However, too frequent dosing of CD40 agonists may result in immune exhaustion [33], which makes this combination of targets challenging in a bispecific setting.

CD40 bispecific antibodies that target tumor stromal proteins is another approach for conditional CD40 activation directed to the TME. One example is bispecific antibodies that target fibroblast activation protein-α (FAP) which is overexpressed on stromal cells in the TME. Two CD40 × FAP bispecific antibodies, MP0317 and RG6189, have been evaluated in clinical studies [49,51,52]. In addition, a preclinical bispecific antibody binding the stroma-expressed target platelet-derived growth factor receptor beta (PDGFRB) and CD40 is under development [53]. By selecting targets on stroma cells, the immune activation is predominantly directed to the stromal areas of tumors, including e.g. fibroblasts or mesenchymal cells. Preclinical studies demonstrated superior anti-tumor effects of the CD40 × FAP bispecific antibodies compared to a CD40 monotargeting antibody in a model with FAP-transfected tumor cells [51]. However, this comparison was made in a tumor model overexpressing FAP, essentially making FAP an artificial highly expressed tumor cell-associated target reminiscent of the Neo-X-Prime approach. Although it was shown that a CD40 × FAP bispecific antibody could possess anti-tumor activity also in a model where FAP was expressed by stromal cells, the anti-tumor effect in this setting was more modest [51]. While stroma-targeting CD40 bispecific antibodies seem to be safe and well tolerated, they do not induce uptake of tumor debris and subsequent cross-priming of tumor-specific T cells as TAA-targeting CD40 bispecific antibodies do [41].

Bispecific antibodies targeting CD40 and receptors expressed on T cells are another therapeutic approach in development. The most advanced bispecific antibody in this class, GEN1042, targets CD40 and 4-1BB and is currently evaluated in a phase 1/2 study [57,65]. This bispecific antibody has been shown to induce conditional activation of both CD40 and 4-1BB [66] and focuses on stimulation of the T cell/DC synapse aiming at reactivating tumor-specific T cells. Data presented with GEN1042 has demonstrated good tolerability and early signs of clinical activity [66,67]. Further, CD40 × PD1 bispecific antibodies are currently in development, and recent preclinical data demonstrated anti-tumor activity superior to the combination of CD40 and PD-1 monotargeting therapies [56].

2.2. Other CD40 targeting modalities in immuno-oncology

The functional consequences of targeting CD40, enhancing upregulation of co-stimulatory molecules and licensing DC to activate CD8⁺ T cells, make CD40-targeting a promising combination partner with cancer vaccines. CD40-targeted vaccines, a form of DC-targeted vaccine, are currently under development with the purpose to both target the antigens to DCs and provide enhance antigen-presentation to T cells. DC-targeting vaccines direct antigen uptake to specific receptors expressed on the cell surface of DCs, including CD40 [68,69]. Data from several phase I clinical trials show that DC targeted vaccines are safe and tolerable in patients [70–72], however, no CD40-targeting vaccine has yet reached the clinic. Compared to other DC receptors, antigen targeting via CD40 has been shown to induce superior T cell responses because of efficient cross-presentation to T cells [73,74].

Direct fusion of antigens to CD40 agonists has been used to direct uptake of antigens to dendritic cells resulting in a very efficient priming of T cells [75–81]. Alternatively, chemical conjugation has been investigated [82–84]. However, these approaches utilize predetermined tumor antigens and are not amendable to a personalized vaccine approach, using neo-antigens specific for an individual patient.

An interesting approach that allows for DC-targeted vaccines to be amenable to personalized use is non-covalent assembly using high affinity peptides [85]. The Adaptable Drug Affinity Conjugate (ADAC) platform enables agonistic CD40 antibodies to deliver antigens to APCs and evoke enhanced T cell responses [86]. The platform is based on non-covalent high-affinity interaction between single-chain variable domains (scFv) fused to the antibody with tagged antigenic peptides. Another approach in development is a CD40 agonistic antibody fused to strongly interacting coiled coil domains that can act as a linkage between the antibody and tumor antigen (Nyesiga et al, manuscript in preparation). In summary, DC-targeting vaccines binding to CD40 has the potential to induce strong anti-tumor immune responses and may be an important component of future personalized vaccine strategies.

CD40 targeted nanoparticles, i.e. nanoparticles with a CD40 binder on the surface, have been used to improve cancer vaccine delivery and efficacy where the tumor antigens can be encapsulated in the nanoparticle and CD40 targeting allow for efficient immune stimulation and delivery to APCs [87–90]. Different forms of nano-engineered platforms have been investigated in this context, including amphiphilic poly(γ-glutamic acid) nanoparticles (γ-PGA NPs) [87], PEGylated unilamellar liposomes [88], poly(lactic-co-glycolic acid) nanoparticles [90], injectable polymer-nanoparticle (PNP) hydrogels APC [91], and lipid nanoparticles [92]. These platforms offer key benefits such as flexibility to fine tune the physicochemical properties of the nano-carriers, possibilities to modify the nanomaterials with targeting molecules and simultaneous encapsulation of immunostimulators and tumor antigens. Improved priming of CD8⁺ T cells, antitumor responses and controlled tumor progression in mouse models have been reported from studies utilizing CD40-targeted nanoparticles [87,88,91]. Further, CD40 targeting nanoparticles can effectively deliver mRNA to DCs both in vitro and in vivo [92]. In summary, nano-engineered platforms in combination with anti-CD40 antibodies have the potential to improve cargo delivery to APCs and induce anti-tumor immune responses.

Gene therapy-based approaches that build on CD40 stimulation have been explored utilizing different viral delivery systems. There are potential advantages with this approach, including the possibility to deliver CD40 agonistic effects directly to the TME, however, the inherent high immunogenicity of the vector makes multiple administrations challenging. Adenovirus-based CD40 ligand gene therapy approaches have generated encouraging results in mice [93–95] and humans [96–99] both by inducing immunogenic cell death and driving efficient T cell responses. In clinical settings, gene therapies that induce CD40-stimulation have been found to be well tolerated with early signs of clinical activity and pharmacodynamic activity in the TME [98,99].

Another approach utilizes a tumor selective adenoviral vector (NG-350A) expressing a fully human agonistic anti-CD40 antibody in the TME [100]. NG-350A was shown to be stable in blood and well-tolerated following intravenous administration. Initial safety and biomarker data from a phase I trial (NCT03852511) indicate that NG-350A can remodel the TME by inducing anti-CD40 antibody production in the TME. Based on these findings, NG-350A is currently being assessed in cancer patients [101,102].

3. CD40 activation of immune cells in cancer

A thorough understanding of the multifaceted roles of CD40 signaling on immune cells, and in particular APCs within the TME, is critical for the development of next-generation CD40 agonists in immuno-oncology. APCs such as DCs, macrophages, and B cells play an important role in immune surveillance, tumor antigen recognition, cross-presentation, recruitment, and activation of other immune cell types. CD40 is expressed on most APCs and CD40 signaling has the potential to increase the tumor-targeting T cell pool, while CD40 signaling in macrophages and B cells promotes a less immunosuppressive TME. Recent technologies for high-dimensional biomarker analysis have shed light on the heterogeneity and plasticity of the APC compartment in various solid tumor types. Here, we summarize recent discoveries of CD40 expression on subsets of APCs and other cell types and the effect of CD40 stimulation on these populations (Figure 3).

Figure 3. Effects of CD40 stimulation on immune cells in the tumor microenvironment. CD40 stimulation on cDC1 and cDC2 license the priming and activation of CD8⁺ and CD4⁺ T cells, respectively. Further, CD40 activation on either of these cell types allows for the transition into mregDcs, which further can sustain CD8⁺ T cells. When acting on tumor-associates macrophages, CD40-targeting therapies favor anti-tumor effects both by shifting the TAM population to favor M1-like macrophages and by increasing the tumoricidal effects of M1 macrophages. CD40 activation on B cells induces the formation of tertiary lymphoid structures and induction of antibody production. Lastly, CD40 stimulation on neutrophils within the TME induces its activation and degranulation, hence promoting anti-tumorigenic effects. cDC: conventional dendritic cell; TAM: tumor-associated macrophage; TLS: tertiary lymphoid structure. Created with BioRender.com.

Several subsets of DCs express CD40 yet the expression varies depending on the activation state and spatial context of the DCs. Several subsets of human tumor-associated DCs have been identified, including the conventional DC (cDC) types 1 and 2 (cDC1 and cDC2), also known as classical DCs. cDC1 is the main population involved in cross-presentation of tumor antigens to CD8⁺ T cells whereas cDC2s are mainly involved in presentation of extracellular antigens to various subsets of CD4⁺ T helper cells [103]. The functional specialization of DC3, pre-DCs/DC5 and monocyte-derived DCs within the human TME has not yet been described, and it is not known how the DC subset diversity is affected by treatment. Apart from the DC subtypes, several studies have identified the mature regulatory DC (mregDCs) phenotype in solid cancers [103]. mregDCs, also called activated DCs, LAMP3⁺ DCs or migratory DCs, lack cell-type specific markers of plasmacytoid DCs (pDCs), cDC1s, cDC2s or DC3s and are characterized by a mature phenotype and the expression of immunoregulatory markers [104–107]. Further, in mice, both cDC1s and cDC2s have been described to contribute to the mregDC population upon tumor cell encounter [108]. Similar findings have been reported in human HPV⁺ tonsillar cancer, where both cDC1 and cDC2-like cells seem to mature into mregDCs [104]. On gene expression level, CD40 has been shown to be expressed on DC3s in oropharyngeal squamous cell carcinoma [109] and on pDCs in pancreatic ductal adenocarcinoma (PDAC), albeit to a lesser extent than cDCs [110,111]. In a reanalysis of public single-cell RNA sequencing (scRNAseq) data sets, Wattenberg et al [111] reported that CD40 was highly expressed by cDCs in PDAC [110], glioblastoma [112], colorectal cancer (CRC) liver metastases [113] and head and neck squamous cell carcinoma (HNSCC) [114]. Additionally, CD40 has been shown to be more highly expressed on cDC1s compared to cDC2s in both CRC liver metastases [111] and treatment naïve CRC [115]. Similar findings have been made in HNSCC, however, with the additional identification of mregDCs showing higher CD40 expression than both cDC1 and cDC2 [111]. In summary, CD40 is expressed widely across DC subtypes with higher expression on cDC1 than cDC2, however, both cDC1 and cDC2 can converge into a mregDC phenotype with even higher CD40 expression. The role of CD40 on DC3 and DC5 remains to be determined.

The essential involvement of DCs in initiating anti-tumor immune responses is well-characterized preclinically [116]. It is well-known that CD40 stimulation results in an up-regulation of co-stimulatory molecules on DCs, such as CD80 and CD86, and production of cytokines which ultimately improves antigen presentation and activation of T cells [1,3,5,20,117]. However, knowledge on the functional consequences of CD40-targeting therapies on different DC subsets beyond cDC1 and cDC2 is limited yet potentially of substantial importance for development of effective CD40-targeting therapies. It has been suggested that CD40 antibodies act mainly on cDC1s and cDC2s within the TME [118] and are able to activate both subtypes [119]. cDC1s are important for effective tumor rejection [28] and analysis of genes induced in cDC1s after CD40 stimulation has revealed a multimodal effect of CD40 activation, including induction of costimulatory ligands for CD8⁺ T cells and promotion of cDC1 survival during the initiation of anti-tumor responses [120]. Additionally, the tumor reduction induced by CD40 targeting agonistic antibodies is lost in mice lacking cDC1s [118], which has been shown to be associated with absence of circulating tumor-specific CD8⁺ T cells [119]. A subsequent study showed that the therapeutic response to anti-CD40 treatment was dependent on preexisting cDC1-primed CD8⁺ T cells [121]. After CD40-targeted therapy in mouse models, the frequency of both cDC1 and cDC2 was reduced whereas mregDCs increased and it has been suggested that mregDCs could mediate activation of preexisting cDC1-primed CD8⁺ T cell clones [121]. Taken together, these data illustrate the intricate interplay between subtypes of DCs within the TME following CD40 stimulation and highlight the importance of continued studies of these subtypes.

Macrophages constitute an important and diverse element of the TME. Traditionally, their diversity has been dichotomized into two polarization states – pro-inflammatory, M1, and immunosuppressive, M2, macrophages. An increased presence of M2 macrophages has been correlated with unfavorable cancer prognosis. Therefore, efforts have been made to therapeutically shift the balance from a M2 to M1 status [122]. However, recent studies using single-cell technologies have revealed that the M1/M2 paradigm does not capture the full spectrum of macrophage heterogeneity within the TME [122]. Ontogenetically, TME macrophages primarily originate from infiltrating monocytes, however a subset of tumor-associated macrophages (TAMs) consists of persisting resident tissue macrophages (RTMs). Importantly, distinct spatial localization patterns have been observed for tumor-associated RTMs and monocyte-derived TAMs across various cancers, with RTMs predominantly located at the tumor periphery and monocyte-derived TAMs infiltrating the tumor core [123]. Due to the large diversity and plasticity of the macrophage population there has been no consensus for nomenclature of subsets of TAMs.

Because of the importance of TAMs within the TME and the widespread expression of CD40 on TAMs [110–112,114], the effects of CD40 targeting therapies on TAMs have gained increasing focus. CD40 stimulation has been shown to favor M1-like human macrophage development in vitro [124], to skew tumor-infiltrated macrophages toward a pro-inflammatory M1 phenotype in mice [111,125] and to sensitize the tumor to chemotherapy [3,4]. Although CD40 stimulation of TAMs in general induces pro-inflammatory M1 macrophage phenotypes, the mechanism behind CD40-mediated skewing of M2 to M1 macrophages is not fully understood and further studies on the effect of CD40 stimulation on different TAM subsets are needed.

B cells express high levels of CD40 and it is well established that CD40 stimulation of B cells plays an important part in the humoral response. The role of B cells in the TME has been highlighted by recent studies demonstrating that presence of tumor-infiltrating B cells and tertiary lymphoid structures (TLS) correlates with better treatment response in different cancer patient populations, indicating that B cells may play an active role also in controlling tumor growth [126–128]. In metastatic melanoma samples, co-occurrence of infiltrating CD8⁺ T cells and CD20⁺ B cells is associated with improved survival outcome, and these tumors display TLS formation [126]. CD40 signaling in B cells results in up-regulation of CD80, CD86 and MHC class II [129], and it has been shown that B cells activated by CD40 can present tumor antigens to T cells and induce anti-tumor immunity [130]. A recent study suggests that CD40 stimulation affects TLS formation [131], which supports further studies to elucidate the role of CD40 stimulation in the context of intratumoral B cells and TLS.

Finally, neutrophils have been described to up-regulate CD40, both in response to CD4⁺ T cell interaction [132] and in response to CD40 stimulation [133]. In addition, CD40 stimulation was shown to result in priming of the NADPH oxidase, suggesting that CD40 interaction can promote ROS production in neutrophils, which in turn can reduce tumor growth [134]. The role of neutrophils in immuno-oncology is increasingly being recognized, and further studies of the role of CD40 stimulation of neutrophils in cancer immunotherapy are warranted.

The first generation CD40 agonists reported dose limiting toxicities including mild to moderate cytokine release syndrome (CRS) and chills, headache, nausea, likely mediated by elevations in IL-6 [16]. Further, moderate transient elevation of liver enzymes likely mediated by CD40 activated Kupfer cells was also reported [16]. However, emerging data with second-generation monospecific CD40 agonists show that with optimal design, CD40 monospecific antibodies can be administered at relevant dose levels in combination with chemotherapy with very limited additional toxicities [6].

4. Combinations and clinical development paths for next-generation CD40 agonists

While the next generation of CD40 agonists may be utilized as single agent therapies, future development in this field will likely be focused on combination therapies. CD40 agonists have been shown to synergize with many different therapies in preclinical models (comprehensively reviewed elsewhere [1,135]). Combination therapies that increase availability of tumor antigens, such as chemotherapies, radiotherapies, antibody drug conjugates, tyrosine kinase inhibitors and cancer vaccines, have strong preclinical and emerging clinical support [4,136–142]. The rationale for these combinations is increasing availability of tumor antigens allowing CD40-induced stimulation of DCs to increase priming of tumor specific T cells. Another approach is to combine CD40 agonists with immune checkpoint inhibitors or other T cell targeting therapies that can enhance the activity and functionality of the tumor specific T cells generated by CD40 agonists [1,5]. Several preclinical studies as well as early clinical data have shown that CD40 agonists in combination with ICI can provide synergistic activity and overcome PD-1 resistance in more inflamed tumors [27,143–145]. Additional combination opportunities include anti-angiogenic treatment to enhance immune cell trafficking and therapies that coactivate myeloid cells [111,146,147].

The two most straight forward opportunities, in our opinion, is combination of next generation CD40 agonists with standard of care chemotherapies in non-inflamed tumor types such as pancreatic cancer or colorectal cancer or combination with ICI in more inflamed tumor types to overcome ICI resistance.

5. Expert opinion

Immuno-oncology has made a remarkable contribution to the treatment of cancer and is now recognized as the fourth pillar of cancer care alongside surgery, radiotherapy, and chemotherapy. Despite the remarkable breakthroughs in the development of immunotherapies over the last decade, less than half of the cancer patients are eligible for immunotherapies such as ICI [148]. Of these eligible patients, less than 15% respond to standard of care immunotherapies.

Thus, there is a high medical need for new therapies that i) can enhance response rates and increase the depth and durability of response in immunotherapy eligible patient populations and ii) can broaden the number of cancer indications where immunotherapies provide clinical benefit. CD40 agonistic therapies provide an opportunity to meet these medical and biological needs. Combining CD40 agonists with standard of care ICI increases the ability to activate tumor specific T cells, potentially overcoming primary and secondary ICI resistance and increase response rates in indications where ICI is currently approved. Further, combining CD40 agonists with chemotherapy is a promising approach in indications where immunotherapies currently are not approved, such as (non-MSI^high) pancreatic cancer and colorectal cancer.

The opportunity to achieve these effects is based on the two-fold mode of action of CD40 agonists by i) activating myeloid cells, such as macrophages, skewing them toward a more tumoricidal phenotype, with the potential to reverse the immunosuppressive TME and increase sensitivity to chemotherapy and ii) activating DCs resulting in priming and expansion of tumor specific T cells with the potential to provide long term benefits to cancer patients.

As highlighted in this review, there are today multiple different next-generation approaches beyond monospecific antibodies to target CD40 in immuno-oncology. Enhancing efficacy should be the main driver, and approaches that maximize the ability of CD40 to both remodel the TME and boost the anti-tumor T cell response have the best chance of providing benefits to the patients. A strong safety profile is also important, both to allow for combination with standard of care therapies without adding toxicities and to allow use also in earlier lines of treatment, e.g. in the neoadjuvant setting. In our opinion, the Neo-X-Prime bsAbs fulfills all these criteria.

In addition to developing the best CD40 agonistic modality, there are other aspects of critical importance to elucidate, such as dosing frequency, combination partner, and dosing regimen in relation to the combination partner. Some reports suggest that too frequent dosing may lead to immune exhaustion [33] but, apart from that, the optimal dosing regimen for CD40 agonists remains to be determined. The optimal dosing schedule may differ when combining CD40 agonists with other therapies.

Promising combination partners for CD40 therapies includes chemotherapy and ICI, but multiple other combinations have shown promise in preclinical models [5]. Chemotherapy treatment results in a massive release of tumor antigens, but also affects the myeloid compartment and the quantity and quality of CD40-expressing cells. This calls for study designs that can balance these effects. A phase 2 study in pancreatic cancer patients (OPTMIZE-1) evaluating a novel dosing regimen geared to maximize the benefits of the CD40 agonist mitazalimab has shown very promising duration of response and survival benefits in combination with mFOLFIRINOX [6].

The indications where CD40 agonists will provide optimal clinical benefit remain to be determined. There is a clear case for CD40 induced T cell priming in non-inflamed tumors, such as pancreatic cancer, in combination with chemotherapy. However, there are also opportunities for CD40 targeting therapies in more inflamed tumor types, such as bladder cancer, where CD40 stimulation can revive ongoing anti-tumor immune responses. Important clues to the activity may be found in the different subtypes of CD40 expressing APCs in the TME. With the advance of single cell sequencing, several new subtypes of myeloid cells have been identified. Improved translational studies on the role of these subsets in response to CD40 stimulation may aid in defining the right patient populations and matching the right indication and combination treatment with the right CD40 targeting approach.

To summarize, next-generation CD40 agonists have the potential to address key biological and medical needs within immuno-oncology. The most promising approaches, e.g. the Neo-X-Prime, effectively target the myeloid cell compartment and enhance the tumor specific T cell compartment.

Article highlights

• CD40 targeting therapies provide an opportunity to enhance response rates and broaden the number of cancer indications where immunotherapies provide clinical benefits by promoting priming of tumor-specific T cells and reverting the suppressive tumor microenvironment.

• Bispecific CD40 targeting antibodies that maximize the ability of CD40 to both remodel the tumor microenvironment and boost the anti-tumor T cell response have the potential to meet key needs in immuno-oncology.

• Enhancing efficacy is the most important driver for development of new CD40-targeting therapies, although ensuring safety is critical to allow for safe dosing at effective dose levels, also in combination with other treatment modalities.

• Bispecific CD40 antibodies dominate the next-generation CD40 therapies, but other treatment modalities, such as CD40 targeting vaccines and nanoparticles as well as gene therapies utilizing CD40 agonists, provide additional opportunities.

• The efficacy of next-generation CD40 agonists depends on the ability to activate the right CD40 expressing immune cell populations. There is a need to improve our understanding of the role of different CD40-expressing immune cell populations and the functional consequences of CD40 stimulation in the tumor microenvironment to facilitate design and development of novel CD40 agonists.

Declaration of interest

P Ellmark, K Enell-Smith and K Hägerbrand are employed by Alligator Bioscience. B Nyesiga and H Andersson are industrial PhD students employed by Alligator Bioscience. M Lindtsedt was visiting professor at Alligator Bioscience 2022–2023. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This work was supported by the Swedish Foundation for Strategic Research (ML: Strategic Mobility, SM21-0042; HA: Industrial PhD student, ID22-0048) and the Knowledge Foundation, Sweden (BN: ComBine Industrial PhD school, 20180114).