Current Situation and Prospect of Adoptive Cellular Immunotherapy for Malignancies

Mechanism of Action and Features

Activated TILs can induce tumor cell apoptosis. TIL therapy uses components of a patient's own immune system to seek out and attack specific cancer cells. TILs are composed of T cells with multiple TCR clones capable of recognizing a range of tumor antigens, and thus, may be superior in addressing tumor heterogeneity compared to other adoptive cell therapies such as CAR-T.^6,7 In 1982, Dr Steven Rosenberg and colleagues at the National Institutes of Health (NIH) were the first to isolate TILs from multiple mouse tumor models. They then demonstrated that co-administration of TILs with IL-2 cured 100% of mice with liver metastases and 50% of mice with lung metastases in an MC38 colon adenocarcinoma model. ⁸

TILs are isolated from fresh tumor tissues via surgery or aspiration. Cytokines, such as IL-2, are then added in the laboratory to allow TILs to proliferate in vitro; TILs are then infused into the patient (Figure 1). The number of TILs and their ability to eliminate cancer cells are then greatly enhanced. TIL production varies from patient to patient, with an average yield of 1.79 × 10⁷ TILs.

OPEN IN VIEWER

Figure 1.

TIL therapy process.

In tumors, the activity of TILs is regulated by a variety of factors. TILs can modulate PD-1 activity in pancreatic cancer receptors. Inhibition of PD-1 signaling enhances the proliferation of pancreatic TILs from tumor fragments and increases the expansion of pancreatic tumor-specific T cells. Meanwhile, 4-1BB, also known as CD137, is mainly expressed on the surface of CD4⁺T and CD8⁺T cells and is an activating immune checkpoint molecule. The combination of 4-1BB and 4-1BBL can activate downstream NF-κB, JNK/SAPK, p38 MAPK, and other pathways, further generating costimulatory signals to induce the activity of CD4⁺ and CD8⁺ T cells and promote the proliferation of T cells. Similarly, the stimulation of 4-1BB with agonistic antibodies favored the expansion of CD8⁺ T cell populations in triple-negative breast cancer (TNBC), pancreatic cancer, and glioblastoma, increased TIL production, and enhanced cytotoxicity.^9–11 Co-stimulatory and co-inhibitory molecules are key players in the activation of the adaptive immune system, regulating the expansion and effector functions of antigen-specific T cells. CTLA4 plays a key role in this interaction, suppressing the immune response to self-antigens. ¹² Niaragh reported that blockade of CTLA-4 induces tumor infiltration by activated lymphocytes regardless of clinical responses in humans. ¹³ These results suggest TIL therapy as a practical anticancer treatment.

Current Status of Clinical Therapy

Currently, among the 408 trials of TIL therapy conducted between 2014 and 2023, 61.76% (252/408) have been opened, 15.69% (64/408) have been completed, and 24.26% (99/408) have either been closed or terminated; 52.45% (214/408) of these trials were in the United States, and 13.48% (55/408) were in China. The immune system, particularly intraepithelial TILs, plays a broad role in controlling tumor growth in nearly all solid tumors. ¹⁴ TIL therapy has primarily been tested as second-line treatment, and melanoma remains the most clinically tested tumor type among followed by NSCLC, ovarian cancer, and head and neck cancers. ¹⁵ TILs play an important role in mediating the response to chemotherapy and improving clinical outcomes in all breast cancer subtypes. TNBC may most likely have >50% lymphocytic infiltration, with the greatest survival benefit for every 10% increase in the number of TILs. ¹⁶ Recent evidence suggests that TILs present in breast cancer before treatment initiation can predict the response to treatment and improve prognosis. ¹⁷ A recent meta-analysis by Hong evaluated the prognostic value of TILs in gynecological cancer. They included 281 articles and showed that there was a significant association between intraepithelial TILs and survival, with an odds ratio of 2.24 for patients without TILs. ¹⁸ The recently published clinical trials of TIL therapy in malignancies are listed in Table 1.

OPEN IN VIEWER

Table 1.

Published Clinical Trials of TILs Treatment for Malignancies.

Completion year	ClinicalTrials.gov	Tumor type	Treatment	Phase	Status	Country
2021	NCT01807182	Melanoma	Aldesleukin, Cyclophosphamide, Fludarabine	Phase 2	Completed	United States
2021	NCT01993719	Melanoma	Aldesleukin, Cyclophosphamide, Fludarabine	Phase 2	Completed	United States
2020	NCT03287674	Ovarian Cancer	Cyclophosphamide, Fludarabine, Ipilimumab, Nivolumab	Phase 1/2	Completed	Denmark
2020	NCT03403634	Colorectal Cancer	Celecoxib, Interferon Alfa-2b	Phase 2	Completed	United States
2018	NCT01820754	NSCLC	Ipilimumab, Paclitaxel, Cisplatin, Carboplatin	Phase 2	Completed	United States
2018	NCT02354690	Melanoma	Vemurafenib, Lymphodepleting chemotherapy	Phase 1/2	Completed	United States
2017	NCT02375984	Melanoma	TILs	Phase 2	Terminated	United States
2017	NCT01814046	Melanoma	Aldesleukin, Cyclophosphamide, Fludarabine	Phase 2	Terminated	United States
2016	NCT01585428	Human Papillomavirus-Associated Cancers	Fludarabine, Cyclophosphamide, Aldesleukin	Phase 2	Completed	United States
2016	NCT02111863	Melanoma	41BB Selected TIL	Phase 2	Terminated	United States
2015	NCT01236573	Melanoma	Cyclophosphamide, Fludarabine	Phase 1/2	Terminated	United States
2013	NCT01468818	Melanoma	Cyclophosphamide, Fludarabine	Phase 2	Terminated	United States
2012	NCT01369875	Melanoma	Cyclophosphamide, Fludarabine	Phase 2	Terminated	United States
2012	NCT00863330	Melanoma (Skin)	TILs	Phase 2	Terminated	United States

The tumor microenvironment in pancreatic cancer is immunosuppressive, with several mechanisms to escape immune surveillance such as co-inhibitory ligands PD-L1 and PD-L2. Moreover, blockade of PD-1 can release CD8⁺ T cells. ¹⁹ Tycko's research found that treatment with the hypomethylating drug decitabine resulted in the production of CD4⁺ and CD8⁺ TILs, resulting in tumor necrosis and slowing tumor growth while increasing expression of PD-1. The best results were obtained when pancreatic ductal adenocarcinoma (PDAC) and anti-PD-1 were combined, which extended the mean survival from 26 days to 54 days. ²⁰ The findings of their study on TIL therapy are promising for future research.

Perspectives of TIL Therapy in Solid Tumors

TIL therapy is a promising immunotherapy for solid tumors that is likely to be implemented in clinical practice. The anti-tumor effects of TILs in vitro and in vivo are greatly enhanced after IL-2 co-culture. IL-2 can enhance the activity and induce proliferation of T cells, which subsequently kill cancer cells. A large dose of IL-2 causes a strong immune response; however, it may also induce organ injury, particularly to the heart, lung, nervous system, kidneys, and other core organs.^21,22 Therefore, to reduce the side effects caused by high-dose IL-2 and improve the in vivo survival rate and function of TILs, it is necessary to actively develop next-generation TIL products. Next-generation TIL therapies are comprised of genetically modified TILs that can knock-out target genes using several techniques, such as CRISPR, to reduce reliance on high doses of IL-2. Unlike PBMCs, TILs have different cellular composition and growth rates. Hence, gene editing that targets TILs may be technically challenging. ²³ Additionally, production of TILs takes time, often more than a month, which may be too long for patients with rapidly developing tumors.

Structure and Development History

One of the most promising approaches in anticancer therapy is CAR-T therapy, which was first proposed by Gross in the late 1980s and has been in development for more than 30 years. ²⁴ The Food and Drug Administration (FDA) approved the first CAR-T therapy product in 2017, which officially announced the industrialization of CAR-T technology, bringing the field into an era of fast-paced and innovative research. ²⁵ CAR-T cells are genetically engineered T cells collected from a patient's own blood to express a CAR on their surface that recognizes specific tumor antigens. CAR is a protein receptor that enables T cells to recognize specific proteins (antigens) on the surface of tumor cells; CAR-T cells can recognize and bind tumor antigens to attack tumor cells. The CAR gene contains sequences that express single-chain variable fragments (ScFvs) of the heavy and light chains of a monoclonal antibody that recognizes a specific antigen. The variable heavy and light chains in ScFvs are linked together by short peptides. In addition to ScFvs, CARs contain a hinge designed to support ScFvs, transmembrane domains, and signaling endodomains that initiate intracellular signaling cascades for antigen recognition.^26,27 Through antibody binding, CAR removes human leukocyte antigen (HLA) restrictions, making CAR independent of antigen presentation and HLA downregulation commonly found in tumors and making CAR applicable to all patients independent of their HLA haplotype. ²⁸ From the perspective of technological development, CAR technology is constantly innovating. Presently, CAR technology has developed up to the fifth generation. ²⁹

First-generation CARs are fusion proteins consisting of an extracellular antigen-binding domain (usually a single-chain variable fragment of an antibody) linked to an intracellular signaling domain (usually the CD3ζ chain of a T-cell receptor) without additional costimulatory domains. ³⁰ In second-generation CARs, costimulatory domains, such as CD28 or CD137, are added to enhance the cellular activity of CAR-T cells to support the expansion and persistence of genetically engineered cells in vivo. ³¹ Third-generation CARs have multiple costimulatory signaling domains, including CD3ζ-CD28-OX40 or CD3ζ-CD28-41BB. This greatly enhances the activity of T cells compared to the second generation.^32,33 Fourth-generation CARs, called T cells redirected for universal cytokine-mediated killing, are cosmic-grade cytokine-mediated killers that are designed to deliver genetically engineered products into tumor tissues after binding to the targeted antigen. As multiple costimulatory domains in third-generation CARs failed to improve the efficacy of CAR-T cells, fourth-generation CARs were based on the second-generation construct. CAR-T cells have been engineered to contain the nuclear factor of the activated T cell-responsive cassette (containing a transgenic cytokine such as IL-12). By inducing the production and release of IL-12, T cell activation was improved, and the approach also successfully avoided systemic toxicity, one of the most common drawbacks of CAR-T therapy.^34,35 A fifth generation of CAR-T cells is currently being developed wherein an additional membrane receptor, such as the IL-2 receptor, is integrated to activate the JAK/STAT pathway in an antigen-dependent manner ³⁶ (Figure 2). One of the most exciting developments is the discovery of switch receptors, making CAR-T cells more controllable than those of previous generations, leading to better safety and wider therapeutic windows.^37,38 The historical timeline of CAR-T cell development is shown in Figure 3.

OPEN IN VIEWER

Figure 2.

Different generations of CAR-T cells.

OPEN IN VIEWER

Figure 3.

Historic timeline of the development of CAR-T cells.

Current Status of Clinical Treatment

CAR-T therapy has shown good targeting, killing, and persistence in clinical trials, providing an alternative for immune cell therapy and showing great development potential and application prospects. Currently, research and development of CAR-T therapy is mainly led by China and the United States. The performance of China is particularly outstanding, and the number of clinical trials in this country has risen rapidly, ranking first in the world. ³⁹ The recently published clinical trials on CAR-T therapy for malignancies are listed in Table 2.

OPEN IN VIEWER

Table 2.

Published Clinical Trials of CAR-T Treatment for Malignancies.

Completion year	ClinicalTrials.gov	Tumor type	Target	Phase	Status	Country
2022	NCT03049449	Lymphomas	CD30	Phase 1	Completed	United States
2021	NCT04160195	B-cell Malignancies, Hodgkin's Lymphoma	CD19, CD20	Phase 1	Terminated	United States
2021	NCT02706392	Malignancies	ROR1	Phase 1	Terminated	United States
2021	NCT03338972	Multiple Myeloma	BCMA	Phase 1	Completed	United States
2021	NCT04097301	Acute Myeloid Leukemia, Multiple Myeloma	CD44	Phase 1/2	Terminated	European Commission
2020	NCT03330834	Lung Cancer	PD-L1	Phase 1	Terminated	China
2020	NCT03289455	Acute Lymphoblastic Leukemia	CD19, CD22	Phase 1/2	Completed	United Kingdom
2020	NCT03958656	Multiple Myeloma	SLAMF7	Phase 1	Completed	United States
2019	NCT03287804	Multiple Myeloma	APRIL	Phase 1/2	Terminated	Netherlands
2019	NCT03019055	Relapsed Refractory B Cell	CD19, CD20	Phase 1	Completed	United States
2019	NCT02215967	Multiple Myeloma	BCMA	Phase 1	Completed	United States
2018	NCT01583686	Metastatic Cancer	Mesothelin	Phase 1/2	Terminated	United States
2018	NCT01454596	Malignant Gliomas	EGFRvIII	Phase 1/2	Completed	United States
2018	NCT02659943	B-cell Malignancies	CD19	Phase 1	Completed	United States
2018	NCT02664363	Malignant Glioma	EGFRvIII	Phase 1	Terminated	United States
2018	NCT02935543	Acute Lymphoblastic Leukemia	CD19	Phase 2	Terminated	United States
2017	NCT02535364	B-cell Acute Lymphoblastic Leukemia	CD19	Phase 2	Terminated	United States
2016	NCT01593696	B Cell Leukemia, Lymphoma	CD19	Phase 1	Completed	United States
2015	NCT00924326	B-cell Lymphoma	CD19	Phase 1/2	Completed	United States
2014	NCT01218867	Metastatic Cancer	VEGFR2	Phase 1/2	Terminated	United States

CAR-T has been successfully used in a variety of tumor types and has achieved remarkable curative effects. Studies have shown that CAR-T therapy is the most effective in treating hematological malignancies. The most important clinical results are from a trial investigating a second-generation autologous CD19-specific CAR-T therapy, starting with significant preliminary clinical results in relapsed chronic lymphocytic leukemia (CLL). ⁴⁰ Subsequent reports showed promising results with CD19 CAR-T therapy in acute lymphoblastic leukemia and diffuse large B-cell lymphoma.^41,42 CAR-T therapy has also shown long-term disease control in cases of hematopoietic cancer. ⁴³

Meanwhile, CAR-T therapy has shown poor performance in solid tumors; however, its implementation in solid tumors is still in its earliest stages. In a variety of solid tumors, most CAR-T research focused on NSCLC. ⁴⁴ A number of clinical trials on CAR-T therapy for NSCLC have been conducted, such as EGFR CAR-T or PSCA CAR-T, but these are still in phase I. ⁴⁵ Furthermore, in another preclinical study, Claudin18.2-redirected CAR-T cells showed promising efficacy against gastric cancer. The overall response and disease control rates were 48.6% and 73.0%, respectively, while the 6-month response rate was 44.8%, showing good efficacy and acceptable safety. ⁴⁶ PDAC extracellular matrix is abundant. The components as well as the production of a highly dynamic and low-vascular tumor microenvironment by the mother cells form a physical barrier that promotes tumor progression and treatment resistance and also hinders the development of CAR-T cells, preventing their detection, transportation, and penetration and limiting their anti-tumor response. ⁴⁷ Although factors, such as THBS1, THBS2, and PEDF, can inhibit the formation of blood vessels reduce blood vessel formation and promote tumor-associated lymph angiogenesis in iCCA, ⁴⁸ the tumor microenvironment remains a significant barrier to CAR-T development, and many target antigens used in cellular immunotherapy, including CEA, CD133HER2, PSCA, MUC1, and MSLN, have been identified and tested in PDACs, including clinical and preclinical,^49–53 with varying degrees of success.

Limitations and Challenges

Although cell-based clinical trials of CAR-T therapy have shown positive results, treatment may also cause serious and potentially life-threatening toxicity.^54–56 The most common serious toxicity is cytokine release syndrome, which is characterized by high fever, sinus tachycardia, hypotension, hypoxia, decreased cardiac function, and other organ dysfunction.^57,58

Antigen escape is one of the most challenging limitations of CAR-T therapy. Although targeting a single antigen can provide high response rates, a large proportion of patients treated with CAR-T therapy have malignant cells that exhibit partial or complete loss of target antigen expression. However, dual-targeting CAR-T cells, such as CD19/CD22 CAR-T cells or HER2 and IL13Ra2, could be used in the future. Both dual targeting CAR-T cells have better anti-tumor responses and reduced immune escape compared to single targeted therapy.^59–61 An additional challenge is that solid tumor antigens recognized by CAR-T cells are often also expressed by normal tissues. Antigen selection is therefore critical not only to ensure therapeutic efficacy but also to limit on-target off-tumor toxicity. In the future, it will be necessary to further develop novel technologies to reduce CAR-T therapy toxicity. Application of CAR-T therapy in solid tumors is limited by cell transportation and infiltration. Therefore, local administration can be adopted to minimize the interaction of CAR-T cells with normal tissues and reduce cytotoxicity.^62,63

Structure and Mechanism of Action

TCR-T cells are created when a gene is inserted into ordinary T cells, resulting in the expression of TCRs that effectively recognize tumor cells. Unlike CAR-T therapy, TCR-T therapy has many advantages. Due to the limited number of membrane proteins in CAR-T therapy, the targets that can be applied to solid tumors are very limited. Meanwhile, TCR-T therapy has a wider target range. Intracellular tumor-associated antigens can be presented in the form of peptides in the major histocompatibility complex (MHC) on the cell surface. TCR-T cells can recognize antigens on the surface and inside of tumor cells through the presentation of MHC, and all antigens that can be presented by MHC can be recognized by TCR-T cells. ⁶⁴

The native TCRs on T cells consist of 4 different T cell antigen receptor polypeptides (α, β, γ, and δ) that form 2 different heterodimers (α:β and γ:δ). Approximately 95% of T cells in peripheral blood are composed of α:β chains, while 5% are composed of γ:δ chains. ⁶⁵ On the surface of T cells, CD3 molecules bind to TCR non-covalently to form a TCR/CD3 receptor complex. When TCR binds to antigen-specific peptide/MHC, Src kinases leukocyte-specific tyrosine kinase and Fyn are activated to phosphorylate ITAM, which activates the Syk family kinase zeta-activated protein 70 kDa, resulting in T cell activation while promoting cytokine secretion, granule secretion, and cell movement and proliferation.^66,67

Four generations of TCR-T therapy have been developed. The first generation of TCR-T therapy is a T cell subset that is directly isolated from a patient and is specifically recognized by tumor antigens, expanded in vitro, and reinfused for treatment. However, due to the extremely small number of such T cell clones and large individual differences, this method was difficult to industrialize. The second generation of TCR-T cells were specifically recognized by the above tumor antigens, obtained its TCR gene sequence, and transduced into peripheral T cells of patients. Therefore, this method makes it possible to industrialize TCR-T therapy. Third-generation TCR-T therapy optimizes the affinity of TCR, enabling it to better recognize tumor antigens, and are then transduced into patient T cells, thereby improving the overall drug ability of TCR-T therapy. Fourth-generation TCR-T therapy is highly specific and targets tumor neoantigens; tumor responsiveness and safety are also greatly improved. ⁶⁸

Current Status of Clinical Treatment

TCR-T therapy has a wider target range and can recognize internal tumor antigens. Additionally, TCR-T therapy has extremely high antigen sensitivity, and each TCR-T cell only needs 1 to 50 antigens to complete the activation process. Hence, TCR-T therapy can recognize low-abundance tumor antigens and better penetrate solid tumors. These advantages make TCR-T therapy more suitable for solid tumors. ⁶⁹

As of September 1, 2022, the search term “TCR” (and synonyms “T Cell Receptor”) in clinicaltrials.gov yielded 671 interventional trials. The first TCR-T trial was conducted in 2004 by Steven Rosenberg of the NIH against the melanoma differentiation antigen gp100 (NCT00085462). Since then, the number of TCR-T trials for different diseases has steadily increased, but most have remained in phase I and II. Most TCR-T trials involved solid cancers (85%) followed by hematological malignancies (9%) and both solid and hematological cancers (2%). There is also a small subset of TCR-T trials for HIV, CMV, and EBV infection (4%). ⁷⁰ The recently published clinical trials on TCR-T therapy are listed in Table 3.

OPEN IN VIEWER

Table 3.

Published Clinical Trials of TCR-T Treatment for Malignancies.

Completion year	ClinicalTrials.gov	Tumor type	Target	MHC	Phase	Status	Country
2021	NCT04476251	Stage IIB-IVA Cervical Cancer	HPV E7	HLA-A*02:01	Phase 1	Terminated	United States
2021	NCT02992743	Advanced Myxoid/ Round Cell Liposarcoma	NY-ESO1	HLA-A*02	Phase 2	Completed	United States
2020	NCT04015336	HPV-Associated Oropharyngeal Cancer	HPV E7	HLA-A*02:01	Phase 2	Terminated	United States
2020	NCT03937791	Vulvar High-Grade Squamous Intraepithelial Lesions	HPV E7	HLA-A*02:01	Phase 2	Terminated	United States
2020	NCT02588612	Non-Small Cell Lung Cancer	NY-ESO1	HLA-A*02	Phase 1	Completed	United States
2020	NCT03168438	Multiple Myeloma	NY-ESO1	HLA-A*02	Phase 1	Terminated	United States
2019	NCT03197025	Vulvar High-Grade Squamous Intraepithelial Lesions	HPV E6	HLA-A*02:01	Phase 1	Terminated	United States
2016	NCT02280811	HPV-Associated Cancers	HPV-16 E6	HLA-A*02	Phase 1/2	Completed	United States

NY-ESO-01 is a testicular cancer antigen that is only expressed in cancer cells; it is considered as the most promising target of TCR-T cells. ⁷¹ Recently, phase I trials of NY-ESO-1-specific TCR-T cells conducted by Hiroshi Shiku et al showed that TCR-T cell infusion exhibited significant tumor response in patients with tumors that express high levels of NY-ESO-1. ⁷² The FDA has approved TCR-T therapy for patients with inoperable or metastatic synovial sarcoma and melanoma. The European Medicines Agency has also approved TCR-T therapy as Priority Medicines.

Highly variable and rapidly proliferating tumor cells often lead to intracellular genetic instability, leading to numerous genetic mutations. ⁷³ Gene mutations and their products that are prevalent in cancer cells are not found in healthy cells and the normal genome, while the expression of non-synonymous mutations can generate tumor-specific antigens, so-called neoantigens. ⁷⁴ KRAS is one of the most common oncogenes in solid tumors, and KRAS mutations are present in approximately 30% of all tumors. KRAS G12D is the most common KRAS mutation in pancreatic cancer, accounting for 41% of all cases. ⁷⁵ Tran et al engineered T cells so that TCR could accurately recognize peptides containing the KRAS G12D mutation without responding to wild-type KRAS. In a patient with advanced metastatic pancreatic cancer, a single infusion of 16.2 × 10⁹ autologous T cells reduced visceral metastases by 72%, and treatment response lasted for 6 months. This finding provides a therapeutic strategy for KRAS G12D-expressing pancreatic cancer and other solid tumors. ⁷⁶

Challenges of TCR-T Therapy

Although TCR-T therapy has shown clinical efficacy, it still faces several challenges. Notably, targeting normal tissue causes immunotoxicity. Although TCR-T cells can target all tumor antigens that can be presented by MHC, the number of targets identified to date with sufficient safety and efficacy is still limited. Therefore, when selecting a suitable target antigen, the primary consideration must be target antigens that are highly expressed in tumors but have low expression or are not expressed in normal tissues. Garcia et al ⁷⁷ developed an alternative strategy to isolate TCR mutants that exhibited high activation signals coupled with low-affinity pMHC binding. Additionally, their model exhibited enhanced target killing potency and undetectable cross-reactivity as well as reduced potential for adverse cross-reactivity. Xu et al ⁷⁸ designed a dual-signal integrator that contains an activating receptor (activator) and a separate inhibitory receptor (blocker). This system employs a blocker targeting a ubiquitously expressed HLA class I antigen to inhibit CAR activation and is designed to activate only when a specific HLA class I antigen is lost, resulting in high sensitivity and specificity. Regarding CAR-T, a key solution to the off-target problem is a genetic safety switch that makes them easier to control once they enter the body, leading to better safety and a wider therapeutic window. ⁷⁹ Correctly pairing transgenic α and β chains is one of the core challenges that hinders the development of TCR-T therapy. ⁸⁰ Inappropriate α/β chain TCR pairing will compete for CD3 complexes, thereby reducing surface expression and signaling of therapeutic TCRs. ⁸¹ One approach to promote correct TCR chain pairing involves the modification of the introduced TCR gene, including the introduction of disulfide bonds, replacement of human constant regions with murine, codon optimization, TCR chain leucine zipper fusion, and single-chain TCR.^82,83 Expression of cancer-associated antigens varies among different cells within a tumor, which allows some tumor cells to evade specific antigen-targeted therapy that leads to treatment resistance in patients receiving immune checkpoint inhibitors ⁸⁴ and adoptive T-cell therapy. ⁸⁵ The simultaneous expression of TCR and co-stimulatory proteins can be achieved by knocking out specific genes while eliminating inhibitory signals. The function of TCR-T cell products can be promoted by preventing T cell dysfunction, inhibiting tumor escape, overcoming limited T cell proliferation, and controlling toxicity. Overcoming these challenges will be keys to its clinical application.

Structure and Mechanism of Action

The success of CAR-T therapy has sparked enthusiasm for genetically modifying NK cells with CARs to enhance their tumor-killing capabilities. NK cells have greater potential as cellular anti-cancer therapy as they may be safer, cheaper, and more rapid-acting. NK cells are potently cytotoxic against tumor cells and are very attractive candidates for next-generation cancer immunotherapy. ⁸⁶ The availability of various sources of NK, such as human cord blood (CB), enhances their potential as therapeutic products for broad clinical scalability. ⁸⁷

CAR-NK therapy uses genetic engineering to add a chimeric antibody to NK cells that allow them to recognize and kill tumor cells. CARs can significantly improve the specificity of NK cells. Similar to the construction of CAR-T cells, CAR-NK cells include an extracellular recognition domain (such as ScFv) to recognize tumor-specific antigens; a transmembrane domain and an intracellular signaling domain (CD3ζ chain) can induce NK cell activation.

Unlike CAR-T cells, CAR-NK cells retain the intrinsic ability to recognize and target tumor cells through their natural receptors. ⁸⁸ Therefore, when CAR-NK targets tumors, the probability of tumor cells escaping killing is greatly reduced. At the same time, CAR-NK cells will not cause immune rejection in the body. Therefore, they do not have the same safety concerns, such as cytokine release syndrome, observed in many CAR-T therapy clinical trials. ⁸⁹ Additionally, NK cells do not require strict HLA matching and have no potential to cause graft-versus-host disease, which is a risk during CAR-T immunotherapy. ⁹⁰

Current Status of Clinical Treatment

CAR-NK cell therapy kills hematological and solid tumor cells in preclinical and clinical trials, demonstrating its potential as an off-the-shelf product with broad clinical applications. Heat shock protein 70 (Hsp70) is highly expressed in 70% of middle and advanced lung cancers. A recent phase II clinical trial for lung cancer showed that NK cells were first activated with Hsp70 in vitro, and NK cells were first recognized by tumor cells in vitro. NK cells are then infused into a patient, allowing them to attack cancer cells in the body. Such a method of stimulating NK cells with antigenic peptides in vitro to enhance specificity has become a research hotspot. A total of 16 patients with stage III NSCLC were enrolled in a clinical trial. One group received traditional combined chemotherapy and radiotherapy, and the other received chemotherapy combined with Hsp70 pre-stimulated NK cell infusion therapy. Results show that Hsp70 pre-stimulated NK cell reinfusion therapy, and when combined with traditional chemoradiotherapy, can double the 1-year survival rate from 33% to 67%; the preliminary results are encouraging. ⁹¹ In highly malignant pancreatic cancer, CAR-NK therapy also showed an ideal therapeutic effect. The umbilical CB-derived engineered CAR-NK cell therapy (PSCA-CAR_s15-NK) developed by Yu et al is highly effective in a mouse model of human metastatic pancreatic cancer, persisting in mice for more than 90 days and significantly prolonging survival without showing treatment-related toxicity. ⁹² CAR-NK therapy offers hope for the treatment of pancreatic cancer.

In addition to solid tumors, CAR-NK therapy also has good efficacy in lymphoma and hematological tumors. CB-NK cells transduced with fourth-generation vectors encoding anti-CD19 CAR and IL-15 induce greater expansion and longer-term persistence in vivo compared to non-transduced NK cells. In a mouse model of lymphoma, although the tumor was not cured, CAR-NK therapy extended the survival of mice. ⁹³ Umbilical CB-derived CAR-NK cell therapy targeting CD19 has also shown significant efficacy in most patients with relapsed/refractory non-Hodgkin's lymphoma and CLL. Eight of the 11 patients (73%) who participated in that study reported a response to treatment, and 7 had complete tumor suppression. The rest of the patients continued to receive post-remission therapy. Importantly, they did not develop cytokine release syndrome or neurotoxicity. ⁹⁴ The recently published and ongoing clinical trials are demonstrated in Table 4.

OPEN IN VIEWER

Table 4.

Published and Ongoing Clinical Trials of CAR-NK Treatment for Malignancies.

Last update year	ClinicalTrials.gov	Tumor type	Target	Phase	Status	Country
2023	NCT05213195	Colorectal Cancer	NKG2D	Phase 1	Recruiting	China
2023	NCT04847466	Head and Neck Cancer	PD-L1	Phase 2	Recruiting	United States
2023	NCT05528341	Solid Tumors	NKG2D	Phase 1	Recruiting	China
2023	NCT05739227	r/r B-cell Hematologic Malignancies	CD19	Phase 1	Recruiting	China
2023	NCT03056339	B Lymphoid Malignancies	CD19	Phase 1/2	Completed	United States
2023	NCT05665075	Acute Myeloid Leukemia	CD33	Phase 1	Recruiting	China
2022	NCT05645601	r/r B-cell Malignancies	CD19	Phase 1	Recruiting	China
2022	NCT05410717	Advanced Solid Tumors	CLDN6	Phase 1/2	Recruiting	China
2022	NCT05194709	Solid Tumors	5T4	Phase 1	Recruiting	China
2022	NCT03692663	Castration-Resistant Prostate Cancer	PSMA	Phase 1	Recruiting	China
2022	NCT05248048	Liver Metastatic Colorectal Cancer	NKG2D	Phase 1	Recruiting	China
2022	NCT05654038	B Cell Hematologic Malignancies	CD19	Phase 1/2	Recruiting	China
2022	NCT05507593	Extensive Stage Small Cell Lung Cancer	DLL3	Phase 1	Recruiting	China
2022	NCT05410041	Relapsed/Refractory B-cell Malignancies	CD19	Phase 1	Recruiting	China
2022	NCT05570188	B Cell Hematologic Malignancies	CD19	Phase 1/2	Withdrawn	China
2022	NCT05472558	B-cell NHL	CD19	Phase 1	Recruiting	China
2022	NCT05008575	Acute Myeloid Leukemia	CD33	Phase 1	Recruiting	China
2021	NCT04796675	B Lymphoid Malignancies	CD19	Phase 1	Recruiting	China
2021	NCT04796688	Hematological Malignancies	CD19	Phase 1	Recruiting	China
2019	NCT03940820	Solid Tumors	ROBO1	Phase 1/2	Unknown	China
2018	NCT03415100	Solid Tumors	NKG2D	Phase 1	Unknown	China

Challenges Ahead

Poor persistence of infused cells in the absence of cytokine support is one of the major drawbacks of adoptive NK cell therapy. While it may be safer, it also limits the effects of NK cell immunotherapy. ⁹⁵ Although exogenous cytokines increase the proliferation and durability of adoptive NK cells, they can also cause side effects, including the growth of suppressive immune subpopulations such as Tregs. ⁹⁶

Rapid homing to the tumor is critical for the efficacy of adoptive cell therapy, a process controlled by chemokines released by tumor cells. However, the efficiency of NK cells homing to tumor sites has been controversial. Related studies have applied various engineering approaches to improve NK cell homing. For example, NK cells are electroporated with mRNA encoding the chemokine receptor CCR7 to increase migration to lymph nodes expressing the chemokine CCL19. ⁹⁷ Additionally, NK cells transduced with a viral vector encoding CXCR2 exhibited better motility against renal cell carcinoma tumors expressing cognate ligands such as CXCL1, CXCL2, CXCL5, CXCL6, and CXCL8. ⁹⁸

Even if the best CAR-NK cells can be created, they will always encounter a suppressive tumor microenvironment, which may be the biggest challenge. ⁹⁹ Much effort has been made to overcoming the negative effects of the tumor microenvironment, such as combining CAR-NK cells with immune checkpoint inhibitors PD-1 ¹⁰⁰ or using HIF-1α-driven CARs that can take advantage of a hypoxic environment. ¹⁰¹

Development and Advantages

Currently, tumor immunotherapy has mainly focused on adaptive immune systems, such as T cells and B cells.^102–104 However, the use of innate immune cells, such as macrophages, in cancer therapy has not been extensively studied. The role of macrophages should be considered in the treatment of solid cancers given their ability to phagocytose, present antigens, and infiltrate the tumor microenvironment. ¹⁰⁵ In addition to the established tumor environment, macrophages are the central effectors and regulators of the innate immune system and are key effectors of cancer therapy based on targeted antibodies. ¹⁰⁶ Macrophages are professional antigen-presenting cells. Additionally, macrophages can also directly kill tumor cells. ¹⁰⁶ Both the activation and inhibition of receptors affect the activity of macrophages. In this process, phagocytosis can also be enhanced by reducing anti-phagocytosis signals. Presently, the most important phagocytosis axis is CD47 on tumor cells and SIRRPα on macrophages. Recent studies have also combined anti-CD47 antibodies with the anti-CD20 antibody rituximab to treat B-cell lymphoma, resulting in a complete remission rate of 36%. ¹⁰⁷

In June 2018, CAR-T cell therapy experts Dr Saar Gill and Dr Michael Klichinsky from the University of Pennsylvania first announced the development of CAR-M therapy to treat tumors, which marked the official establishment of CAR-M therapy. ¹⁰⁸ CAR-modified macrophages are considered a promising cell type. CAR-M requires the extraction of macrophages and the introduction of CARs into them through genetic engineering to ultimately achieve tumor killing. Similar to CAR-T and CAR-NK cells, CAR-M cells consist of extracellular signaling domains, transmembrane regions, and intracellular activation signaling domains that recognize specific tumor antigens. ¹⁰⁹ Compared with immune cells such as T cells and NK cells, macrophages may infiltrate tumors in an immunosuppressive microenvironment more easily, providing more opportunities for tumor immunotherapy. ¹¹⁰

Unlike CAR-T cells, CAR-M cells have unique advantages. T cells cannot enter the tumor environment due to the physical barrier formed by the matrix surrounding tumor cells, whereas macrophages can visibly infiltrate the tumor environment. Tumor-associated macrophages (TAM) play important roles in tumor invasion, metastasis, immunosuppression, and angiogenesis. CAR-M therapy can reduce the proportion of TAM, affect the cell phenotype of TAM, and have a positive effect on tumor treatment. ¹⁰⁹ In addition to phagocytosing tumor cells, CAR-M therapy also promotes antigen presentation and enhances T cell killing.^105,111

Current Status of Clinical Treatment

CAR-M therapy was first reported in 2020, with most of the studies being in the scientific stage. The FDA has granted Fast Track designation to HER2-targeted CAR-M CT-0508 for the treatment of patients with solid tumors. Researchers from the University of Pennsylvania used an anti-HER2 CAR-M containing the CD3-ζ intracellular domain. In 2 ectopic mouse models of solid tumor transplantation, a single injection of anti-HER2-CAR-M reduced tumor burden and prolonged mouse survival. In a humanized mouse model, HER2-CAR-M also converted M2 macrophages into M1 macrophages, induced an inflammatory tumor microenvironment, and enhanced the anti-tumor cytotoxicity of T cells. ¹¹² There are 2 other clinical trials based on CAR-M strategies that have been approved and are underway. One is the drug candidate CT-0508, which treats patients with relapsed/refractory HER2-overexpressing tumors with an anti-HER2 CAR-M (phase I clinical trial). The other is MCY-M11, which uses mRNA to transfect PBMCs to express CARs targeting mesothelin (including CAR-M) for the treatment of patients with relapsed/refractory ovarian cancer and peritoneal mesothelioma; it is currently recruiting volunteers for phase I clinical trials.

Significant attention was given to the choice of the signaling domain when designing CAR-M cells. CAR-M derived from mouse cytoplasmic structures of phagocytic receptors, such as Megf10, FcRγ, Bai1, and MerTK, were designed. ¹¹³ Primary murine macrophages expressing FcRγ or Megf10-based CARs exhibit antigen-specific phagocytic capacity. When screened in the RAW264.7 cell line, CAR-M with the MerTK activation domain exhibited the greatest tumor cytotoxicity. However, studies have reported that although anti-CCR7 MerTK CAR-M performed well in theory, Morrissey's anti-CD19 CARs with the same cytoplasmic domain was unable to bind antigen-functionalized beads, suggesting that generating new CAR-M architectures requires careful optimization and functional evaluation. ¹¹³

Challenges Ahead

CAR-M therapy does have obvious advantages in solid tumors, but it is still in its “infancy.” To clinically apply CAR-M therapy for solid tumors, several difficulties need to be overcome. Macrophages are the body's first line of defense against cancer cells and viruses due to their ability to phagocytose pathogens. However, primary human macrophages are not easily transfected by viral vectors commonly used in genetic manipulation. ¹¹⁴ As a result, a viral vector cannot infect macrophages, which is a major challenge limiting the development of CAR-M therapy.

Mass expansion of genetically engineered macrophages is also a concern. Unlike T cells, which can be cloned and expanded in large numbers in the laboratory, macrophages have far lower differentiation and proliferation capabilities ¹¹⁵ and hardly expand in vitro. The maximum number of macrophages that can be obtained from a patient at one time is limited, which may severely limit the efficacy of CAR-M therapy.

The therapeutic effect of CAR-M therapy is also affected by the migration of macrophages in vivo. After CAR-M cells are infused into a patient, exogenous macrophages first pass through the lungs, and most remain in the liver; only 30% are recruited to the tumor site within 5 days.^112,113 Thus, it is not conducive for cancer treatment.

Therefore, in follow-up clinical trials, it is necessary to continuously observe and test the efficacy and possible challenges of CAR-M therapy.