Targeting the CD47-SIRPalpha checkpoint in multiple myeloma

Abstract

Immune evasion is a hallmark of cancer and there is mounting evidence that the tumor microenvironment (TME) plays a role in the pathogenesis of haematologic malignancies as well as treatment resistance. Macrophages play a central role in anti-tumor immunity, and dysregulation of macrophage mediated phagocytosis has recently emerged as a key player in blood cancers. The integrin associated protein CD47 is expressed in a variety of cancers and interacts with its ligand, signal regulatory protein α (SIRPα) expressed on macrophages, resulting in down regulation of macrophage-mediated phagocytosis. CD47 is highly expressed in various cancers including multiple myeloma (MM). It is therefore postulated that blockade of the CD47-SIRPα immune checkpoint has the potential to ‘re-awaken’ macrophage mediated phagocytosis of MM plasma cells. In this review, we provide our perspective on the key pre-clinical data supporting the CD47-SIRPα axis as a therapeutic target in MM. We subsequently discuss the ongoing clinical trials which may provide the basis for future clinical translation of these agents. We also highlight key gaps in our knowledge of macrophage biology in MM which need to be addressed by future research. Finally, we present potential future directions for translational research and personalized application of macrophage-based immunotherapy in MM.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12672-025-03312-6.

Introduction

Multiple myeloma (MM) is the second most common hematologic malignancy worldwide, characterized by the clonal expansion of malignant plasma cells (PCs) in the bone marrow (BM) [1–3]. Despite the introduction of several targeted therapies, MM remains incurable due to the development of treatment resistance in most patients. Immune dysregulation is a hallmark of cancer, and in MM, it plays a critical role in treatment resistance [2, 4, 5]. MM PCs interact with various components of the tumor microenvironment (TME) to promote their survival and proliferation. Dysfunctional natural killer (NK) cells, T cells, and dendritic cells contribute to immune evasion, while myeloid-derived suppressor cells further dampen anti-tumor immunity [5–8]. While impaired T and NK cell responses are linked to poor outcomes, tumor-associated macrophages (TAMs) have recently emerged as key players in MM [9, 10].

Macrophages, essential components of the innate immune system, are involved in processes such as phagocytosis, antigen presentation, and cytokine secretion [11]. In the MM TME, TAMs are critical mediators of tumor progression [12, 13]. These macrophages are thought to be predominantly of the anti-inflammatory subtype, promoting tumor survival and immune suppression [9, 14, 15]. The interaction between TAMs and malignant plasma cells is therefore a key area of interest in MM research.

Strategy for literature search

A comprehensive literature search was performed for published articles on targeting the CD47-SIRPα checkpoint in multiple myeloma (MM) using single or combination therapies including monoclonal antibodies, fusion proteins and other novel therapeutics. The electronic search was conducted using the databases PubMed, Embase, Google Scholar, Web of Science and Scopus. Key search terms included ‘multiple myeloma’, ‘CD47’ and ‘SIRPα’, and the time period of the search spanned from the date of inception till January 2025. Articles describing pre-clinical data, as well as observational and interventional clinical studies were scrutinized, given the novelty of the field, we did not exclude any category of publication from our search. Information regarding relevant ongoing clinical trials was obtained from ClinicalTrials.Gov. Abstracts from international conferences including the American Society of Haematology (ASH) and American Society of Clinical Oncology Annual Meetings were screened for relevant unpublished work.

The CD47-signal regulatory protein α axis in multiple myeloma

Signal regulatory protein alpha (SIRPα), a member of the SIRP family of molecules, was first identified in 1997 [16] and is expressed on myeloid cells such as macrophages, monocytes, dendritic cells and granulocytes [17]. It was subsequently elucidated that CD47, a 47 kDa transmembrane protein ubiquitously expressed on the surface of many cell types, is a ligand of SIRPα [18]. The interaction between CD47 and SIRPα initiates a signaling cascade which culminates in the phosphorylation of immunoreceptor tyrosine-based inhibitory motifs (ITIMs) and deactivation of myosin IIA, a crucial step for inhibiting phagocytosis [19]. The CD47-SIRPα interaction therefore inhibits phagocytic activity, serving as a “don’t eat me” signal to the macrophages [20, 21]. This “macrophage check-point” is thought to be a fundamental mechanism facilitating immune escape by downregulating phagocytosis of malignant cells by TAMs [22]. Indeed, overexpression of CD47 is associated with poor prognosis in a variety of cancers such as esophageal [23], breast [24] and gastric [25] carcinomas, as well as acute myeloid leukemia [26] and non-Hodgkin lymphoma [27].

In MM, CD47 plays a critical role in malignant transformation, with significant upregulation of CD47 observed during the progression from monoclonal gammopathy of undetermined significance (MGUS) to MM [28]. M plasma cells also exhibit markedly higher CD47 expression compared to other cell types in the tumor microenvironment (TME), suggesting that CD47 may be a promising therapeutic target to promote macrophage-mediated phagocytosis of malignant plasma cells [15].

There is an array of studies investigating inhibition of the CD47-SIRPα immune checkpoint through various strategies such as anti-CD47 antibodies, anti-SIRPα antibodies, and soluble SIRPα peptides [29], with the use of anti-CD47 antibodies being the best characterized to date. These agents have displayed efficacy in inducing phagocytosis of tumor cells in both in vitro and animal models of various hematologic and solid tumors [27, 30, 31]. While most of the phase I/II clinical trials evaluating anti-CD47 antibodies have focused on other hematologic malignancies [32], the promising results gleaned from these trials suggest strong potential for further exploration of this strategy in MM. Figure 1 illustrates the biology of the CD47-SIRPα checkpoint and the impact of its therapeutic targeting. Figure 1 illustrates the biology of the CD47: SIRPα checkpoint and the impact of its therapeutic targeting.

Fig. 1.

An overview of the potential factors influencing the CD47-SIRPalpha checkpoint and the downstream effects of checkpoint inhibition in multiple myeloma. TME tumor microenvironment

Clinical studies evaluating CD47 expression in multiple myeloma

Gene expression profiling (GEP) studies on patients with PC dyscrasias have revealed overexpression of CD47 in clonal PCs compared to those from normal controls [33]. In fact, in as many as 65% of MM cells, CD47 expression was at least two-fold higher than that in normal PCs from healthy donor bone marrow [30].CD47 mRNA levels have also been shown to be higher in MM plasma cells compared to the other cells found in the TME [30, 33].

It is noteworthy that CD47 expression based on GEP [15, 33] and real-time reverse transcription polymerase chain reaction [28] increases during the progression from MGUS to MM. Specifically, the median expression level of CD47 mRNA increased approximately 10-fold with progression from MGUS to MM [28], suggesting an important role for the CD47:SIRPalpha checkpoint in the evolution of PC dyscrasias. It is noteworthy that these studies did not compare CD47 expression between high and low risk MGUS or smouldering myeloma, which would be an important analysis to be conducted in future studies.

Importantly, CD47 was found to be highly expressed in newly diagnosed MM patients, although this was not correlated with the stage of disease or the cytogenetic risk groups [15]. It is noteworthy that the expression of CD47 was demonstrated not only at the mRNA level based on publicly available GEP data sets, but also at the cell surface protein level by flow cytometry [15]. The cell surface expression being of particular relevance to therapeutically targeting the check point.

CD47 gene expression was shown to be increased in relapsed compared to newly diagnosed MM [34] and was associated with inferior clinical outcomes in relapsed MM [35]. There have however been no studies studying cell surface CD47 expression at relapse, or in extramedullary disease. This would be important in elucidating the role of CD47 expression at relapse and in the development of independence from the BM microenvironment.

Tumor cells are known to communicate with the TME via a variety of mechanisms and macrophages have been shown to be a key component of this interplay in solid tumors [36]. None of the studies published so far correlated CD47 expression on the PCs with SIRPα expression on the TAMs of the same patients. This would be of particular interest in evaluating whether MM PCs can influence SIRPα expression on TAMs, and whether CD47 itself plays a role in this intracellular communication.

Impact of CD47 expression on response to therapy in multiple myeloma

The first demonstration of the role of CD47 in treatment response was based on immunohistochemical (IHC) analysis of CD47 expression in BM trephine biopsies from 74 newly diagnosed MM patients treated with vincristine, adriamycin and dexamethasone induction followed by autologous stem cell transplant [34]. It was found that high CD47 expression was associated with shorter median progression-free survival and overall survival, suggesting for the first time that the macrophage check point plays a role in treatment resistance.

More recently, the impact of the CD47: SIRPα check point on response to immunotherapy has become an area of interest, with anti-CD38 monoclonal antibodies now being used as first line therapy [37]. Kambhapathi et al. sought to study the role of TAMs and CD47 expression in the response to daratumumab (dara)-based therapy in relapsed/refractory (RR) MM [38]. They demonstrated by IHC that while total CD47 expression level was not altered at relapse following data treatment, expression pattern of CD47 was altered, with a trend of decreased membrane and augmented cytosolic expression at relapse after dara therapy. Although this study attempted to elucidate the role of TAMs in treatment response, their use of a single IHC marker (CD68) limits the analysis.

High dimensional immune profiling using flow cytometry and single cell omics have since shed light on specific subsets of monocytes and macrophages which may play a role in the response to daratumumab based therapy. Specifically, intermediate and classical monocytes were enriched in the peripheral blood of progressors [39] while monocytes with an interferon response signature were detected in the bone marrow of patients with suboptimal responses in a separate study [40]. Given the heterogeneity and plasticity of tumor associated monocytes/macrophages, further analyses are required to more accurately define these subsets. It is also crucial to establish the mechanism by which they exert their pro-tumor function, be it by failure of macrophage mediated phagocytosis or modulation of other immune effector cells. Importantly, SIRPα expression on these subsets was not described in any of the published studies. Nevertheless, these data provide the basis for future studies to explore CD47 and SIRPα in the context of immunotherapy treated MM patients.

Pre-clinical studies evaluating the CD47-SIRPα axis as a therapeutic target

Given the vital importance of the CD47-SIRPα axis in the perpetuation of various malignancies, The use of monoclonal antibodies, recombinant peptide agonists and miRNA therapeutics targeting this checkpoint are being investigated in preclinical models [41]. Early studies in a murine model transplanted with the KPMM2 human MM cell line found remarkable anti-tumor effects of targeting CD47. Treatment with a mouse monoclonal single-chain antibody fragment targeting CD47 attenuated weight loss, delayed the occurrence of hind leg paralysis from spinal cord compression, and significantly prolonged median survival when compared to the control [42]. Similar results have since been reported by other groups [15, 30] Notably, anti-CD47 antibody treatment was found to significantly inhibit tumor growth in a bortezomib-resistant MM murine patient derived xenograft model derived from a patient with an MM related pleural effusion [43]. These pre-clinical data support the possibility of anti CD47 therapy in bortezomib resistant MM as well as extramedullary disease.

The anti-CD47 monoclonal antibody B6H12 was found to potentiate in-vitro phagocytosis of MM cell lines RPMI 8226, NCI H929, KMS18 and MM1.S, as well as patient-derived MM xenografts [30]. In RPMI 8226 xenotransplantation models, B6H12 treatment led to tumor regression in 42 out of 57 mice. Importantly, it was found that CD47 blockade alleviated bone resorption in human bone-bearing mice. Osteoclastogenesis is a cell-contact dependent process which leads to up-regulation of thrombospondin-1 (TSP-1) and plays a pivotal role in the developmental of lytic bone disease in MM [44]. Disruption of CD47-TSP-1 interaction by TSP-1 blocking antibodies or down-regulation of CD47 on tumour cells has been found to abrogate tumour-induced osteoclast formation [44]. Taken together, these results suggest that anti-CD47 therapy is effective in targeting MM PCs via macrophage mediated phagocytosis, in addition to alleviating MM related bone damage.

Sun et al. studied the anti-human CD47 monoclonal antibody Vx1000R in a 2D and 3D-tissue engineered bone marrow model (3DTEBM) [15]. The efficacy of Vx1000R in tissue engineered bone marrow models without co-cultured murine BM macrophages (BMMs) was compared to the efficacy in bone marrow models with co-cultured BMMs. It was found that Vx1000R induced macrophage-mediated phagocytosis of MM cells as early as 4 h following administration. Real-time live confocal imaging demonstrated macrophage-mediated phagocytosis as well as changes in cellular motility following treatment with the antibody. Importantly, Vx1000R did not result in MM cell killing in the 2D culture that was not co-cultured with BMMs, an observation postulated by the authors to be due to the 2D model lacking the complexity to accurately replicate the complex TME of MM. This lends support to the importance of the TME and the utility of 3D cultures to study macrophage-directed therapies.

Following previous in vivo studies demonstrating efficacy of anti-CD47 therapy in leukemia [45], lymphoma [46] and MM [47] models, Tseng et al. sought to investigate whether an adaptive immune response is generated against cancer following anti-CD47 mAb treatment [48]. The immunodeficient mice used to establish xenograft models for haematological malignancies lacked T, B and NK cells making them unsuitable as a model to study the interconnections between the adaptive immune system and anti-CD47 therapy. The colon cancer cell line DLD1 was therefore utilized for this purpose through in-vitro T-cell priming assays as well as a murine model. It was found that in addition to inducing phagocytosis of the tumor cells, targeting CD47 using the mouse antihuman anti-CD47 antibody B6H12 (IgG1) also primed a CD8 + T-cell mediated anti-tumor response and simultaneously downregulated the immunosuppressive response mediated by T-regulatory cells. The involvement of both innate and acquired immune systems in the mechanism of anti-CD47 therapy results in a robust anti-tumor immune response. This raises the plausibility of combination treatment with adoptive T-cell therapy or T-cell engaging antibodies to augment the adaptive immune response against tumor antigens [48].

AO-176 is a humanized IgG2 monoclonal antibody targeting CD47 being investigated in preclinical models. AO-176 possesses various favourable pharmacodynamic characteristics including preferential binding of tumor cells compared to normal cells, direct tumor cell killing, and activation of innate immune mechanisms via induction of damage-associated molecular patterns (DAMPs) [49, 50]. Evaluation of AO-176 in a subcutaneous xenograft model of MM showed significant anti-tumor response and prolongation of survival [49]. Further studies evaluating changes in the host immune system of murine models before and after treatment with CD47 targeting therapeutics may shed more light on their mechanism of action. This would also delineate differences in the immunomodulatory properties between the various agents targeting this checkpoint.

The dysregulated expression of several micro RNAs (miRNA) such as miR-15a and miR-16 has been characterized and is hypothesized to influence CD47 gene-expression in MM cells [51]. It has been found that miR-155 plays a tumor suppressor role and down-regulates CD47 expression [34]. Rastgoo et al. demonstrated that administration of synthetic miR-155 in a xenograft MM mouse model impaired tumor growth and prolonged survival by promoting macrophage mediated phagocytosis. It was also identified that miR-155 resulted in re-sensitizing of resistant MM cells to bortezomib, raising the plausibility of combining miRNA therapeutics targeting the CD47-SIRPα axis with established MM therapies in future clinical trials.

Therapeutic strategies combining anti-CD47 with other forms of immunotherapy, most notably with anti-CD38 monoclonal antibodies have also been explored in vitro. It has been found that treatment with anti-CD47 amplified PC killing mediated by dara [52]. The same study established the importance of monocytes (characterised based on CD14 and CD16 expression) to dara-mediated PC killing in vitro. It was also found that monocytes expressing CD16 were required for anti-CD47 antibody-mediated in vitro killing of PCs, suggesting that the synergistic effects of anti-CD38 and anti-CD47 therapies are facilitated by monocytes. Further studies using high dimensional profiling are required to better characterise the monocyte/macrophage subsets involved.

Novel strategies involving SIRPα/Fc fusion protein antibodies and monoclonal antibodies targeting SIRPα have demonstrated promising preliminary pre-clinical activity in the context of solid cancers and other haemtologic malignancies. TTI-621 is a fusion protein developed from the CD47 binding domain of human SIRPα linked to the Fc region of human IgG1 which has been shown to enhance macrophage phagocytosis in a non-Hodgkin lymphoma model [53]. SIRP-1 and SIRP-2 are novel anti-SIRP antibodies which have been evaluated in B cell lymphoma, colorectal adenocarcinoma, endometrial carcinoma and ovarian carcinoma cell lines [50]. It was found that SIRP-1 and SIRP-2 were efficacious in disrupting CD47-SIRPα interactions and induced phagocytosis of a range of cancer cell lines. Phagocytosis was also found to be augmented when combined with tumor-opsonizing antibodies, including a highly differentiated anti-CD47 antibody AO-176. These data provide the basis for future studies evaluating these agents in MM.

Clinical trials of therapeutics targeting the CD47-SIRPα checkpoint in multiple myeloma

The encouraging results of in vitro studies have spurred hopes of potential bedside translation of agents targeting the CD47-SIRPα axis. Table 1 summarizes the landscape of currently ongoing clinical trials evaluating CD47-SIRPα targeted therapy in MM.

Table 1.

Clinical trials evaluating checkpoint Inhibition of the CD47-SIRPα axis in multiple myeloma

Clinical Trials.gov Identifier	Drug name	Mechanism of action	Study design and description	Current status
NCT04892446	Hu5F9-G4 (magrolimab)	Human IgG4 monoclonal antibody against CD47	Multicenter, phase II study evaluating the efficacy of magrolimab dual therapy with daratumumab, combination therapy with pomalidomide and dexamethasone, and dual therapy with carfilzomib, in patients with relapsed/refractory MM.	Ongoing
NCT04445701	AO-176	Humanized IgG2 monoclonal antibody targeting CD47	Open-label, multi-center, dose-escalation phase I/II clinical trials evaluating efficacy of combination therapy of AO-176, dexamethasone and bortezomib in patients with relapsed/refractory MM.	Ongoing
NCT05139225	Maplirpacept (TTI-622)	Fusion protein of CD47-binding domain of SIRPα linked to Fc region of human IgG4	Open-label, phase I study evaluating TTI-622 in combination with daratumumab in patients with relapsed/refractory multiple myeloma	Ongoing
NCT05567887	Maplirpacept (TTI-622)	Fusion protein of CD47-binding domain of SIRPα linked to Fc region of human IgG4	Open-label, phase I study in patients with relapsed/refractory MM to assess safety and tolerability of Maplirpacept monotherapy	Ongoing
NCT05896774	Maplirpacept (TTI-622)	Fusion protein of CD47-binding domain of SIRPα linked to Fc region of human IgG4	Open-label, phase I study in patients with relapsed/refractory MM to evaluate safety and tolerability of Maplirpacept monotherapy.	Ongoing
NCT03530683	Maplirpacept (TTI-622)	Fusion protein of CD47-binding domain of SIRPα linked to Fc region of human IgG4	Open-label, multi-center phase I study evaluating the safety and efficacy of maplirpacept monotherapy against maplirpacept/carfilzomib/dexamethasone triple therapy and maplirpacept//carfilzomib/dexamethasone combination therapy in patients with relapsed/refractory MM.	Ongoing
NCT05675449	Maplirpacept (TTI-622)	Fusion protein of CD47-binding domain of SIRPα linked to Fc region of human IgG4	Open-label, phase Ib study comparing elranatamab/maplirpacept dual-therapy to elranatamab/carfilzomib/dexamethasone combination therapy in patients with relapsed/refractory MM.	Ongoing
NCT04895410	Lemzoparlimab	Novel, differentiated CD47 antibody targeting a distinct CD47 epitope preferential to tumour cells	Phase 1b, dose escalation and exapansion study of Lemzoparlimab with or without dexamethasone and in combination with anti-myeloma regimens for the treatment of patients with relapsed/refractory MM.
NCT02663518	Ontorpacept (TTI-621)	Fusion protein of CD47-binding domain of SIRPα linked to Fc region of human IgG1	Open-label, phase Ia/Ib trial evaluating Ontorpacept monotherapy in patients with relapsed/refractory MM.	Terminated for administrative reasons

Hu5F9-G4 (magrolimab) is a first-in-class humanized monoclonal antibody against CD47 [54, 55]. Magrolimab was engineered with a human IgG4 isotype which is inefficient at recruiting Fc-dependent effector functions, reducing toxic effects on healthy CD47-expressing cells [55]. Magrolimab was initially evaluated in combination with azacitidine in myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) [56]. While early phase data were promising, the development of this drug has been halted in myeloid malignancies due to futility and increased toxicity seen in the phase III ENHANCE 2 and ENHANCE 3 clinical trials [57].

A multicenter, phase II study evaluating the efficacy of magrolimab in combination with either dara, pomalidomide, dexamethasone, or carfilzomib, dexamethasone in patients with RR MM is currently ongoing [58] [NCT04892446]. The various combination therapies will first be tested through a safety run-in phase where adverse effects, dose-limiting toxicities and laboratory abnormalities will be determined. Subsequently, the combination therapies will be evaluated in the dose-expansion cohorts (n = 27 assigned to each combination therapy) to determine the clinical efficacy of magrolimab in RR MM. This trial will provide valuable information on the clinical synergy of anti CD47 antibodies with existing MM therapeutics. However, given the safety concerns arising from the MDS and AML trials there is uncertainty regarding the further development of this drug in hematologic malignancies.

Given robust preclinical data (described in the previous section), the safety and efficacy of the monoclonal antibody AO-176 is currently being evaluated in an open-label, multi-center, dose-escalation phase I/II clinical trial for RR MM patients (NCT04445701). The phase II cohorts of the study will assess the efficacy of combining AO-176 with dexamethasone and bortezomib. The end points of the study include the overall response rate, duration of response, progression-free survival, and overall survival. The combination of this agent with monoclonal antibodies and second-generation proteasome inhibitors would be of significant interest and should be evaluated in future trials.

Maplirpacept (TTI-622) is a soluble recombinant fusion protein created by directly linking the sequences encoding the CD47 binding domain of SIRPα with the Fc domain of human IgG4 [59]. This agent is currently under clinical development for the treatment of various haematologic malignancies including lymphoma and AML [59]. Maplirpacept is currently being investigated as monotherapy and in combination with existing MM therapeutics in various phase Ib/II trials. The phase I trials evaluating maplirpacept monotherapy in RR MM (NCT05139225) (NCT05567887) (NCT05896774) are summarized in Table 1. Another ongoing phase I study is evaluating maplirpacept monotherapy as well as maplirpacept/carfilzomib dexamethasone and maplirpacept/isatuximab/carfilzomib/dexamethasone combination therapies (NCT03530683). This trial would generate important data on the optimal combinations to be evaluated in future phase 2 and 3 trials.

Ontorpacept (TTI-621) is a recombinant fusion protein created by linking the CD47 binding domain of SIRPα with the Fc domain of human IgG1. A phase I trial (NCT02663518) was conducted to investigate its safety and clinical efficacy in the treatment of various haematologic malignancies, with early results displaying promising safety and activity in lymphoma [60]. In 2022, the trial was terminated for administrative reasons and the results from the MM arm of the study remain unpublished.

The ongoing MagnetisMM-20 trial has been designed to compare the safety, tolerability and clinical efficacy of elranatamab (a bispecific T-cell engager targeting B-cell maturation antigen) and maplirpacept dual-therapy to elranatamab/carfilzomib/dexamethasone for RR MM (NCT05675449). Given the preclinical data demonstrating augmentation of T-cell responses generated by CD47 blockade, it is plausible that there may be synergy between bispecific T-cell engagers and CD47 blockade. The results of this trial would therefore be of significant importance to the design of future combination therapies incorporating these agents. Lemzoparlimab is a novel, differentiated CD47 antibody targeting a CD47 epitope unique to tumour cells [61]. A phase 1b, dose escalation and expansion study of lemzoparlimab with or without dexamethasone and in combination with anti-myeloma regimens is currently under way (NCT04895410). The toxicity data for this agent would be of particular interest given its higher selectivity for plasma cells and potentially less impact on normal host cells expressing CD47.

The recent development of CD47-CD38 bi-specific antibodies represents an innovative method to block CD47 specifically on plasma cells while sparing normal cells and potentially minimizing toxicities [62, 63]. Early phase clinical trials are currently ongoing (NCT05427812). Preliminary results from this ongoing phase 1/2 study involving 26 patients reveals treatment-related adverse effects such as anemia, neutropenia, thrombocytopenia and cytokine release syndrome [64]. Due to occurrence of cytokine release syndrome from the initial doses studied, a priming dose was introduced in subsequent cohorts which was found to decrease the incidence of cytokine release syndrome. So far, the preliminary results demonstrate manageable toxicity and this trial continues to enroll subjects in dose escalation.

Chimeric antigen receptor (CAR)-T cell therapy is a revolutionary new pillar in oncological therapeutics and has exhibited remarkable efficacy and durable clinical responses in a variety of cancers [65]. CARs are engineered synthetic receptors which function to redirect lymphocytes, most commonly T cells, to recognize and eliminate cells expressing a specific target antigen [66]. B-cell maturation antigen (BCMA) is selectively over-expressed during the malignant transformation of plasma cells and has emerged as an ideal target for the treatment of MM [67]. The FDA has approved Carvykti and Abecma, BCMA-directed CAR-T cells for the treatment of RRMM, yielding high remission rates of up to 72–97% [68]. However, downregulation or loss of BCMA expression has been observed in several MM patients after BCMA-directed CAR-T therapy, which mediates resistance [69]. A novel BCMA/CD47-directed CAR-T therapy has been devised which showed in vitro and in vivo efficacy against MM cells and prolonged survival of tumour-engrafted mice in vivo [70]. These promising results warrant further investigation in human subjects.

Critical view and discussion

Although we have made great strides in our understanding of the CD47-SIRPα checkpoint in MM, there remain several critical unanswered questions to be addressed in future studies. While increased CD47 expression has been found to correlate with progression from MGUS to MM, the question of how CD47 expression changes at relapse, and whether it plays a greater role in the resistance to some therapies than others remain to be answered. These knowledge gaps could be addressed by assessing CD47 expression in patients enrolled in prospective trials of novel therapies.

It is also likely that there exists significant intra patient heterogeneity in CD47 expression at both a spatial and temporal level. Studies evaluating CD47 expression at serial time points in the course of the disease would be of great value in addressing this question. Variation of CD47 expression during clonal evolution of MM is also an important challenge [71], this is particularly the case for MM stem cells. To date, there is still a lack of consensus on the definition of the MM stem cell [72]. Furthermore, there is a dearth of research on the expression of CD47 in MM stem cells, which may be largely hampered by the challenge of accurately defining them [72]. Moving forward, there is a need to better characterize the immune phenotype of MM stem cells, which may unravel potential therapeutic vulnerabilities. Understanding CD47 expression in the context of plasma cell heterogeneity will be crucial to better understand which sub-populations of patients with MM might respond best to anti-CD47 therapy.

The role of CD47 expression in extramedullary MM (EMD) is of particular interest given the clinical challenges faced in treating patients with EMD [73]. Novel therapeutic strategies are urgently required in this subset of patients, and a deeper understanding of their genomic and immunologic landscape is of vital importance [74]. Indeed, it is reasonable to hypothesize that some genomic abnormalities underlying MM maybe associated with CD47 expression and de regulation of TAMs [75]. A deeper understanding of these interactions may pave the way for innovative strategies using genomically targeted therapy in combination with CD47 directed immunotherapy for EMD [76].

It is noteworthy that the bulk of published research focuses on PC CD47 expression while SIRPα expression in myeloma TAMs (an equally important component of the checkpoint) has not been studied in detail. This maybe at least partly due to the challenge of characterizing TAMs in MM which are a heterogenous and dynamic cell type [9]. High dimensional profiling of myeloma TAMs, in conjunction with assessment of their SIRPα expression will be required to address this important knowledge gap.

While SIRPα is a key ligand for CD47 in the context of cancer immunology, it is possible that less well-known ligands may also be relevant. Thrombospondin-1 (TSP1) and TSP2, which are also ligands of CD47, are found at elevated levels in BM cultures from MM patients compared with healthy donors [28]. The role of these ligands in treatment resistance remains unknown and the plausibility of therapeutic strategies targeting these proteins is intriguing. In addition to future studies evaluating the interaction between CD47 and thrombospondins, dual blockade of CD47 and PD-L1, targeting glutaminyl cyclase (an enzyme modifier of the CD47-SIRPα axis) and nanoparticle delivery methods for checkpoint blockade, are all fascinating avenues for future research [77–80]. Chimeric antigen receptor (CAR) T-cells targeting CD47 have shown promising pre-clinical activity in solid tumors [81] and as discussed in Sect. 7, are now being explored in MM. This would be another important avenue to explore among the CD47 directed therapeutics for MM. Potential novel avenues for targeting CD47 are summarized in Fig. 2.

Fig. 2.

Emerging strategies for targeting the CD47-SIRPα axis in multiple myeloma. Novel approaches under investigation include combining CD47 blockade with existing immunotherapies (e.g., monoclonal antibodies, proteasome inhibitors, and immunomodulatory drugs), bispecific T-cell engagers (BiTEs), chimeric antigen receptor (CAR) T-cell therapies, and immune checkpoint inhibitors such as PD-1/PD-L1 blockers. Additional innovative strategies involve nanoparticle-based delivery systems and genomically guided combination therapies. Together, these approaches aim to enhance anti-myeloma immune responses while mitigating toxicity and overcoming therapeutic resistance. BiTE bispecific T-cell Engager, IMiD immunomodulatory drug

The toxicities associated with CD47 directed therapeutics will also be a crucial area of study in the ongoing clinical trials. A full understanding of the potential adverse effects and their severity would be vital to inform the appropriate monitoring and dose adjustments for patients in future clinical trials. Furthermore, identifying biomarkers to predict these toxicities would also be of great interest to enable stratification of patients for these trials. It is encouraging that various modalities are already under investigation to mitigate some of the toxicities of CD47 blocking therapies [80].

While targeting the CD47-SIRPα checkpoint in MM is still in its early stages of clinical development, this is a promising addition to the expanding armamentarium of immunotherapeutics for MM. We propose that a personalized approach to the application of these therapies will be key to achieving optimal outcomes and ongoing translational research will be vital in bringing the field forward.

Author contributions

Conceptualization, T.M.S. and S.M.; writing-original draft preparation, T.M.S. and S.M.; writing-review and editing, T.M.S., W.J.C., H.L. and S.M.; supervision S.M. All authors have read and agreed to the published version of the manuscript.

Funding

SDM is supported by the Singapore National Medical Research Council Transition Award. WJC is supported by the Singapore National Medical Research Council STaR investigatorship.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The authors declare that ethics approval, consent to participate and consent to publish are not applicable for this work.

Competing interests

The authors declare no competing interests.