Cancer is one of the most lethal diseases that has been treated since the existence of human beings. Approximately 19.3 million new patients and nearly 10.0 million cancer deaths were reported worldwide in 2020 (1). In the past few decades, before new therapies were discovered, a combination of conventional treatments such as surgical resection, chemo- and radiation therapies were first-line treatments for many types of primary cancers. Although conventional treatment methods are effective in the early stages of cancer, these treatments are non-selectivity and can cause adverse effects (2). For instance, lymphedema, which is the accumulation of lymph fluid in tissues, can be caused by damage to lymph nodes or blood vessels resulting from surgery or radiation therapy (3). Chemotherapy drugs can cause neutropenia by targeting fast-growing cells such as healthy white blood cells (4). Therefore, studies on a variety of therapeutic interventions for improving the selectivity of conventional cancer treatment are in progress. On that note, numerous anticancer therapies and reagents have emerged and stem cell-mediated therapies using adult stem cells (ASCs) are considered attractive vehicles for delivering anticancer agents (Fig. 1). ASCs, such as mesenchymal stem cells (MSCs), neural stem cells (NSCs), and hematopoietic stem cells (HSCs) are undifferentiated populations that exist in our bodies, and have tumor tropic effect and the potential to partially replace conventional treatments (5).

An important characteristic of adult stem cells is their outstanding migratory and homing ability (6). The migratory ability of ASCs was initially elucidated in a xenograft mouse model with brain tumors (7). The exact mechanism of ASCs tumor tropism is not fully elucidated, but several experimental evidence indicates that various chemokines such as monocyte chemoattractant protein-1, stromal cell-derived factor 1, and stem cell factor-1 are involved (8-11). The efficacy of NSC tropism has been reported in gliomas and other malignancies. It has been demonstrated that NSCs preferentially trace metastatic tumor foci in various mice organs. Intravenous injected NSCs or MSCs can migrate to metastatic nodes of breast cancer in the lungs (12-16). Because of these abilities, adult stem cells have been actively studied as intermediate carriers of anticancer substances that track and kill cancer from the past to the present.

Prodrug-enzyme therapy

Utilizing their migration ability to tumor sites, MSCs and NSCs are genetically engineered to express therapeutic genes. The suicide gene cytosine deaminase (CD), found in Escherichia coli or yeast, converts the prodrug 5-fluorocytosine (5-FC) to the widely used cytotoxic anticancer drug 5-fluorouracil (5-FU) (17-19), which subsequently converts to the activate metabolites, inhibiting RNA and DNA synthesis and functioning of thymidylate synthase in the tumor (18). CD expressing human MSCs, such as adipose tissue-derived MSCs and a clonal immortalized NSC HB1.F3.CD, are both genetically engineered to express the CD suicide gene (20, 21). CD-NSCs have been especially studied in malignant gliomas as a first-generation enzyme/prodrug system to overcome challenges associated with the delivery of anticancer compounds and the replacement of damaged tissues (22, 23). Research using yeast CD gene; yeast cytosine deaminase::uracil phosphoribosyltransferase (yCD::UPRT) has been studied and applied in MSCs (24). Numerous in vitro and in vivo studies have extensively demonstrated the safety and therapeutic efficacy of expressing yCD::UPRT and E. coli CD gene (25-27). Additionally, a clinical trial using CD and 5-FC to treat high-grade gliomas has also been applied and completed (NCT01172964 and NCT04657315) (28).

Likewise, carboxylesterases are enzymes that alter the pro-drug CPT-11 (irinotecan) to its active form SN-38 (7-ethyl-10-hydroxycamptothecin), a robust inhibitor of topoisomerase I (29-31). The pro-drug CPT-11 is currently being used in combination with other anti-cancer drugs for neuroblastoma and has progressed well in phase I clinical trials (32). The adenovirus-transduced HB1.F3.CD (human NSCs) with human enzyme is around 70-fold more efficient at converting CPT-11 to SN-38, than the endogenous wild-type human isoform of CEs in the liver (33, 34). The herpes simplex virus thymidine kinase (HSV-TK) is an enzyme that is widely studied for another suicide gene strategy, using the pro-drug ganciclovir (GCV) (35, 36). Applying the same transduced HB1.F3.CD, HSV-TK phosphorylates GCV into a monophosphate active form, which has the potential to effectively eliminate cancer (37).

Oncolytic virotherapy

Oncolytic viruses (OVs) preferentially target, replicate in, and eliminate cancer cells, rather than other normal cells (38). OVs can induce both innate and adaptive immunity (39). These characteristics have given new directions for anti-cancer therapy. Several oncolytic viruses have been tested in tumor-bearing animal models and have garnered major scientific interest. Viral transduction of MSCs is frequently performed in adeno-, lenti-, and retro-viruses (40). The major obstacle of oncolytic virotherapy is viral destruction by host immunity. Several studies have shown that anti-viral destruction is mediated by endogenous interferon signaling (41). To overcome this limitation, MSC-mediated oncolytic virotherapy has been considered due to their tumor-tropic ability towards tumor foci and immunosuppressive functions (42, 43). MSCs produce several soluble factors and enzymes, such as cyclooxygenase pro-inflammatory cytokine interferon-γ (43). Previous studies have reported that MSCs specifically suppress immune cells, including the activity of T cells, B cells, and maturation of dendritic cells. These features provide robust shelters to safely deliver OVs to tumor sites (44). There has been a lot of research to improve the therapeutic efficacy of OVs. For example, Yoon AR et al. inserted Wnt-signaling inhibiting decoy acceptor sequences into an adenovirus, targeting hepatocellular carcinoma (HCC). This resulted in highly activating the Wnt signaling pathway and effectively destroying HCC. Additionally, reports indicate that MSC-mediated virotherapy prevents unpredictable side effects caused by MSCs due to the cells being lysed by adenovirus replication in vivo (45).

NSC-mediated virotherapy has been studied especially in gliomas and ovarian cancer. Aboody et al. reported that immortalized HB1.F3.CD21 (engineered neural stem cell line) in non-tumorigenic as well as enhancing tumor foci, inhibited the replication of adenovirus in glioma and ovarian cancer models. This oncolytic virus-loaded NSC model is currently in a clinical trial of glioma patients (46). In addition, injection of NSCs with oncolytic adenovirus (NSC-CRAd-S-pk7) successfully distributes the tumor area and allows the virus to deeper penetrate the tumor periphery (47).

Extracellular vesicle-mediated anti-cancer therapy

Extracellular vesicles (EV) are heterogeneous lipid bilayer particles that are produced by many types of cells including stem cells (48). Exosomes are extracellular membrane-bound vesicles ranging from 50-100 nm in diameter, and comprising protein, DNA, and RNA secreted by cells (49). As versatile therapeutic cargo, exosomes have several benefits. For instance, exosomes are involved in various biological processes such as antigen presentation, angiogenesis, and transferring horizontal miRNA (50). Alvarez-Erviti L et al. reported that in a mouse model, systemic injection of dendritic cell-derived exosome effectively delivered siRNA to the brain (51).

Unformulated antisense oligonucleotide (ASO) inhibitors have proven anticancer efficacy in multiple preclinical cancer models (52). ASO AZD9150 and CpG-STAT3ASO conjugates target the inhibition of oncogenic signal transducer and activator of transcription 3 (STAT3), which is an essential promoter of numerous human cancers (53). Adamus T et al. assessed the clinical efficacy of NSC-derived exosomes in the glioma as cargo for the delivery of oligonucleotides. Peritumorally administrated NSCs producing exosomes encapsulated CpG-STAT3ASO have demonstrated that NSC-mediated delivery significantly elevates the percentage of CpG-STAT3ASO and effectively reduces tumor growth, compared to the administration of oligonucleotides alone (54).

The previously mentioned CD/5-FC system has also been used as a therapeutic reagent of the exosome. It has been reported that MSCs engineered to express the CD gene released exosomes contained CD mRNA. The exosomes were internalized by cancer cells and triggers apoptosis due to the conversion of prodrug 5-FC to active drug 5-FU (55). MSC-EVs carrying therapeutic substances such as miRNA were also shown to have anti-cancer effects. In a previous study, it was shown that miRNA-148-3p transferred by human umbilical cord-derived MSC induce apoptosis by regulating tripartite motif 59 in MDA-MB-231 breast cancer xenografts (56).

Cytokine mediated therapy

Cytokines, small molecules normally produced by immune cells, are involved in the regulation of immune responses, and comprise interleukins, chemokines, tumor necrosis factors, and lymphokines (57). In the mid-1990s, a high dose of interleukin 2 was found to be an effective anti-cancer agent, leading to the commencement of cytokine-based therapy (58). Since then, a variety of cytokines have been evaluated for anti-cancer efficacy. However, systemic administration of immunomodulating cytokines demonstrated several shortcomings. Because of the short half-lives of cytokines, administration of high doses of cytokines can cause adverse side effects, including cardiac problems. To prevent unfavorable side effects, ASCs are an appropriate model for delivering cytokines precisely to the tumor environment. IFN-β-inserted MSCs were the first cytokine-engineered MSCs, which inhibited the growth of melanoma in vivo. Also, it was reported for the first time, that MSCs were positioned into the tumor and produced large quantities of IFN-β (59).

Subsequent studies confirmed that diverse cytokines and chemokines, such as the interleukin and interferon families, act as tumoricidal reagents when inserted in MSCs. Interleukin 7, interleukin 2, interleukin 12 (IL-12), interleukin 15, and interleukin 18 are reported as immunoregulatory cellular cytokines against cancer (60). Viral transduced MSCs of IL-12 could migrate to the tumor site, where the MSCs locally express the IL-12 gradually without disturbing other systems, thereby minimizing the inherent adverse side effects (61). MSCs overexpressing tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) showed a 6-fold increase in the ability to kill tumor cells, as compared to the non-modified MSCs (12). MSCs expressing the death receptors TRAIL-R1 and TRAIL-R2 resulted in TRAIL-induced apoptosis of tumor cells expressing the TRAIL death receptors (62). Additionally, a clinical study is being conducted on the therapeutic efficacy of TRAIL for patients (NCT03298763).

Nanoparticle-loaded ASCs anti-cancer therapy

The use of nanoparticle (NP) carriers as a delivery system for selective anti-cancer agents attenuates unfavorable side effects and improves chemotherapeutic pharmacologic properties by altering the biodistribution and pharmacokinetics (63). ASCs such as MSCs and NSCs are considered appropriate vehicles for delivering therapeutic NPs while overcoming drawbacks of traditional chemotherapy, such as suboptimal penetration and retention. Various cationic polymers that destruct tumors chemically or physically as ASC-NP conjugated or in an internalized form have been widely studied (64). For instance, gold nanorods (AuNRs) improve several drawbacks, including damage to the surrounding healthy tissue due to the complex tumor geometries of photothermal therapy, i.e., heat generation when a tumor is exposed to near-infrared radiation (65). Positioning the AuNRs within the tumor tissue converts the 810 nm nonharmful laser light into thermal energy, raising the local temperature to 28-60°C and leading to tumor resorption (66). To overcome this positioning issue, NSCs were selected to deliver AuNRs to the tumor foci. NSC-AuNR conjugates resulted in enhanced tumor surrounding the ability of AuNRs, resulting in improved anti-cancer efficacy, compared to administration of AuNRs alone (67). Mesoporous silica nanoparticles can transport the drug to the tumor site effectively, because of their surface pores. The anti-cancer drug can be released by certain stimuli such as pH, temperature, magnetic fields, and ultrasound (68, 69). It has been reported that mesoporous silica-loaded MSCs release doxorubicin and induce apoptosis of cancer cells when the MSCs are stimulated by ultrasound (70).

HSCs engineered with chimeric antigen receptors for anticancer therapy

HSCs are promising candidates for anticancer therapy, especially leukemia, through the transduction of genes that encode chimeric antigen receptors (CARs) or T cell receptors (TCRs). CARs and TCRs recognize tumor-associated antigens directly, in a major histocompatibility complex (MHC)-independent manner (71). Although these methods have yielded unprecedented competency in B cell acute lymphoblastic leukemia, several challenges still need to be overcome, such as reduced T cell persistence and activity. Besides, the use of mature T cells has a possibility for mispairing, i.e., a fusion of the introduced transgenic TCR chains with endogenous chains. Gene-editing technologies have enabled the elimination of the endogenous TCR and class-1 MHC-I to minimize graft-versus-host disease (GvHD) and allograft rejection (72). Another approach is the transplantation of HSCs. Using TCR- or CAR-transduced HSCs against cancers (especially leukemia) provides some advantages, which simultaneously resolves the challenges of both T cell persistence and TCR mismatching (73). First, the TCR- or CAR-transduced HSCs will constantly produce T-lymphocyte progenitors after transplantation, which thereafter undergo the natural T cell maturity process, developing potential immunity of the recipient. Next, unfavorable cytotoxicity of engineered TCR due to mispairing from transduced mature T cells is potently suppressed by HSCs that have not yet been rearranged. Giannoni F et al. reported that engraftment of transduced HSCs enhances the T cells expressing TCR recognizing tumor-associated antigen, which exerts specific anti-tumor activity (74).

HSC-platelet conjugation-mediated therapy

Hu et al. reported that monoclonal anti-programmed death-1 antibodies (aPD-1) are secreted from the HSC-platelet-aPD-1 (S-P-aPD-1) cellular complex due to platelet activation. The immune checkpoint inhibitor aPD-1 blocks programmed death-1 (PD-1) that are expressed in various immune cells and disturbs the immunosuppressing ability of cancer through PD-1 ligands (PD-L1) (75). S-P-aPD-1 effectively promotes the delivery of aPD-1, and leukemia-specific T cells induce tumor growth suppression in the bone marrow, which is the site of acute myeloid leukemia (76).

ASCs are attractive models and give new insight into anti-cancer therapy, but the application of ASCs will not completely replace previous traditional treatments such as surgery or radiation therapy. Stem cell therapy for cancer may be an attractive choice appropriate for patients diagnosed with metastasized malignancy, who have fewer treatment options than patients in the early stages. The Discovery of the tumor tropism of ASCs has led to the development of varied anti-cancer approaches. Of the various methods reported, engineered stem cell-mediated therapies have shown anti-cancer efficacy not only in various primary cancers, but also in metastatic malignancies (77). As a therapeutic agent, engineered ASC-secreted materials such as oncolytic viruses, cytokines, and proteins exert anti-cancer activity, and subsequently induce the shrinkage of tumor volume via cancer cell death. Although the outcome of ASC-mediated anti-cancer therapy is remarkable, several limitations and concerns about safety and shortcomings still remain. As with other anti-cancer therapies, a single treatment of ASC therapies does not guarantee complete eradication of the tumor and prevent tumor recurrence. Moreover, stem cells that migrate to the tumor site must remain only long enough to secrete the therapeutic agent, but the permanent alive immortalized ASCs should be avoided due to the possibility of them becoming tumorigenic. Portnow J et al. reported this based on first-in-human autopsy data, which stated that only infinitesimal NSCs remain longer than the required period of prodrug conversion. Additionally, it was reassuring to note that NSCs did not seem to undergo division (Proliferating cell nuclear antigen negative) and were positioned as single cells in the tumor (28). Another obstacle to cell-mediated therapy is unequal biodistribution in the tumor foci after stem cell administration. Stem cells trace tumor sites through leaky vessels of the tumor. Depending on the size and type of the tumors, varying degrees of blood vessel leakiness cause disruptions in the biodistribution and dose-response correlation. Moreover, engraftment of allogeneic immune cells and stem cells has the potential to possibly cause infections or GvHD (78, 79). To elucidate these concerns, the increasing number of preclinical and clinical researchers is aiding in developing and standardizing clinical trials. Advanced findings may guide investigators in the next phase of stem cell therapies through cutting-edge technologies, and a better understanding of cancer physiology.

This work was supported by the Basic Science Research Program (2020R1A2C2006060) and the Global Research and Development Center (GRDC) Program (2017K1A4A3014959) through the National Research Foundation (NRF) of Korea, funded by the Ministry of Science and ICT. In addition, this work was also supported by the Basic Research Lab Program (2022R1A4A1025557) through the National Research Foundation (NRF) of Korea, funded by the Ministry of Science and ICT.

The authors have no conflicting interests.