Skip to main content
NCBI home page
Journal of Translational Medicine logoLink to Journal of Translational Medicine
. 2025 Apr 24;23:473. doi: 10.1186/s12967-025-06365-x

MUC18-Directed chimeric antigen receptor T cells for the treatment of mucosal melanoma

Fenghao Zhang 1,#, Haizhen Du 1,#, Kaiping Liu 1,#, Qian Guo 1, Mengmeng Liang 2, Jing Shi 3, Shi Feng 2, Ting He 2, Xin-an Lu 2, Yanfang Tang 3, Lihua Wang 2, Qiaozhen Li 2, Xun Meng 3, Shu-Hui Liu 3,✉, Yanping Ding 2,✉, Yan Kong 1,✉
PMCID: PMC12020318  PMID: 40275251

Abstract

Purpose

Mucosal melanoma, a highly aggressive form of skin cancer, remains challenging to manage due to the lack of effective therapies. Mucin 18 (MUC18) is overexpressed in both primary and metastatic lesions of melanoma but rarely in normal tissues. The expression profile makes MUC18 a potential target for development of therapeutic antibodies or chimeric antigen receptor-T (CAR-T) cell therapy. This study aims to generate an effective CAR-T targeting MUC18-positive melanoma and evaluate its preclinical antitumor activity.

Experimental design

A humanized anti-MUC18 single chain antibody fragment (scFv) was used to construct CAR-T with various designs of the hinge, transmembrane, co-stimulatory, and CD3ζ domains. The antitumor efficacy of MUC18 CAR-T cells was assessed in vitro, in MUC18-positive primary and rechallenged xenograft models, as well as in patient-derived xenograft (PDX) models of human mucosal melanoma.

Results

The humanized scFv selectively bound to MUC18 with high affinity. Various MUC18 CAR-T cells specifically killed MUC18-positive melanoma cells and could proliferate as a result of exposure to antigen. Among them, CAR-T cells containing an IgG4-derived hinge domain and a CD28 co-stimulatory domain demonstrated superior antitumor efficiency. Robust tumor regression and CAR-T cell expansion were observed in multiple MUC18-positive xenograft models after treatment with the IgG4 hinge and CD28 empowered CAR-T cells.

Conclusions

This study demonstrated the development of a novel CAR-T therapy for mucosal melanoma, MUC18 CAR-T, that showed strong potency in tumor eradication and inhibition of tumor relapse. This candidate CAR-T therapy could provide a promising strategy for the treatment of the refractory melanoma.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-025-06365-x.

Keywords: MUC18, CAR-T, Mucosal melanoma, Preclinical efficacy

Introduction

Mucosal melanoma is a rare but aggressive malignancy originating from melanocytes [1]. The average five-year survival of mucosal melanoma is 26.8% [2]. This subtype of melanoma accounts for 25% of melanoma cases in Asian but only 1.4% in Caucasian [3, 4]. Accumulating evidences show that mucosal melanoma has distinct genomic landscape compared with that in cutaneous melanoma [5, 6], including low frequency of BRAFV600 mutations and low tumor mutational burden. Thus, mucosal melanoma normally exhibits intrinsic resistance to molecular targeted therapy and immune checkpoint inhibitors. In contrast, these therapies have been proven successful in the treatment of UV-exposed cutaneous melanoma [7–10]. To date, therapeutic strategies for mucosal melanoma are very limited, and chemotherapy still accounts for the main therapeutic option. Exploring novel targets and corresponding therapies are needed for mucosal melanoma.

In recent years, adoptive cell immunotherapy (ACT) has gained special attention to treat refractory or relapsed cancers [11]. Tumor-infiltrating lymphocyte (TIL) therapy has recently been reported to improve the progression-free survival compared with ipilimumab in patients with unresectable stage IIIC or IV cutaneous melanoma, suggesting that ACT is a potential therapeutic strategy for advanced cutaneous melanoma [12]. However, the effectiveness of TIL in mucosal melanoma remains obscure. Among other ACT strategies, chimeric antigen receptor-T (CAR-T) cell therapy has been proven effective for hematologic malignancies and certain types of solid tumors [13–17], raising an exciting possibility that mucosal melanoma can become curable towards CAR-T cells. So far, several types of CAR-T cells have been reported mostly in preclinical studies to treat metastatic melanoma with candidate targets including VEGFR2, GD2, cMet, CD70, gp100, NY-ESO-1, CD20, IL13Rα2, and B7H3 [18–28], but limited clinical evidences are available. A key challenge for CAR-T therapy in mucosal melanoma lies in selection of appropriate antigen with steady expression in tumors and limited off-target toxicity.

Mucin 18 (MUC18), a member of the cell adhesion molecule family, has been implicated in the pathogenesis of advanced melanoma, with high expression levels observed in both primary and metastatic lesions [29–32]. In the human context, MUC18 exhibits abundant expression in embryonic tissues, contrasting with its minimal presence in adult normal tissues [33]. Distinct from other immunoglobulin-like cell adhesion molecules (Ig-CAMs), MUC18 mediates adhesion through specific ligand interactions rather than homophilic binding, thereby fulfilling dual roles in maintaining cellular mechanical integrity and transducing extracellular matrix signals intracellularly. The repertoire of MUC18’s extracellular matrix ligands encompasses laminin, galectin, and S100A8/A9, which, upon engagement, activate the PI3K/Akt, NF-κB, and MAPK signaling cascades, thereby promoting oncogenic proliferation and migration. Moreover, MUC18 possesses intrinsic tyrosine kinase activity, enabling it to act as a standalone cell surface signaling molecule, transducing signals from Wnt, FGF4, VEGF-C, and Netrin-1, or functioning as a co-receptor for VEGF-A and PDGFR-β.

The low expression of MUC18 in adult human tissues and its high expression in tumor tissues make it an ideal target for various anti-tumor drugs. The therapeutic potential of MUC18 targeting in oncology has been the subject of several investigative studies. Mohammadi et al. identified a high-affinity MUC18-targeting single-chain variable fragment (scFv) that demonstrated inhibitory effects on the migration and invasion of breast cancer cell lines in vitro [34]. Feng et al. generated a MUC18-targeted monoclonal antibody (JM1-24-3) through live cell immunization and assessed its anti-melanoma efficacy both in vitro and in vivo [35]. Zhang et al. evaluated the efficacy of a MUC18 antibody (AA98), derived from human umbilical vein endothelial cell (HUVEC) immunization, in a murine orthotopic uveal melanoma model, revealing significant suppression of tumor growth and angiogenesis [36, 37]. An additional study reported a neutralizing monoclonal antibody that inhibits tumor growth and angiogenesis by blocking the interaction between soluble CD146 and angiomotin, consequently abrogating the activation of the downstream c-Myc pathway [38]. In the realm of nanomaterial-based MUC18 targeting, Liu et al. developed polydopamine nanoparticles (PDA NPs) conjugated to MUC18, which, in conjunction with mild photothermal therapy, effectively inhibited the migration of melanoma and breast cancer cells in vitro [39]. Yu et al., employing a similar approach, engineered MUC18-targeted gold nanorods to augment tumor endothelial cell permeability [40].

In our previous work, we analyze the expression of MUC18 in 97 melanoma samples including cutaneous, acral, mucosal, uveal, and meatastatic melanoma with unknown origin. We identified robust overexpression of MUC18 in all types of melanoma [41]. We also performed a high-throughput screening of an antibody repertoire using mucosal melanoma cell lines, we identified MUC18 targeted monoclonal antibody pAb253 as the sole antibody capable of exerting significant cytotoxic effects on mucosal melanoma cell lines [41]. These findings underscore the potential of MUC18 as a therapeutic target for mucosal melanoma, positioning it as a promising antigen for the development of CAR-T cell therapies.

Materials and methods

Cell lines

A375 cell line (ATCC, Cat#CRL-1619) and HEK293T cell line (ATCC, Cat#CRL-3216) were cultured in DMEM medium (Hyclone, Cat#sh30022.01) supplemented with 10% fetal bovine serum (FBS) (Bioind, Cat#04-001-1C04001-500). GAK cell line was purchased from JCRB cell bank in 2016 and was cultured in F-12 K medium (Gibco, Cat#21127022) supplemented with 20% FBS. HMV-II cell line was obtained from Sigma in 2015 and cultured in RPMI 1640 medium (Hyclone, Cat#sh30809.01) supplemented with 10% FBS. All cell lines were authenticated by short tandem repeat (STR) analysis according to the suppliers’ information.

A375-Luc+ cell was generated by lentiviral transduction of A375 cells with firefly luciferase and GFP genes. A375-MUC18-KO cells cultured in DMEM medium were generated by CRISPR-Cas9 deletion of MUC18 in A375 cells. CRISPR single guide RNA (sgRNA) targeting MUC18 were designed from the website (https://zlab.bio/guide-design-resources) and synthesized by Biomed. The sgRNA sequence was cloned into the PX458 plasmid vector with Cas9 sequence (Addgene Plasmid, #48138). One of five sgRNA (CTGCTGCTGTCCTCGCGTCG) targeting MUC18 was selected for further experiments after validation for highest deletion efficiency. A375 cells were electroporated using the electroporator (Lonza) following program FF-120.

Anti-MUC18 ScFv affinity detection

The binding affinity of scFv derived from pAb253 for human MUC18 was determined by surface plasmon resonance (SPR). Affinity analysis was carried out using a Biacore 8 K (Cytiva, Cat#29337763). Briefly, the mouse anti-human IgG antibody was immobilized on sensor surface of CM5 chip by amine coupling. Anti-MUC18 scFv-Fc was diluted to 2 ug/mL and flow over sensor surface in a running buffer at a flow rate of 10 µL/min. Human MUC18 protein was diluted by double ratio with running buffer. The association and dissociation were measured for 300s. In each case, sensor surface was regenerated by repeated washing with 10 mM HCl and 10 mM NaOH for 30 s at the rate of 30 µL/min. Each binding curve was corrected by subtracting the signal collected from negative control flow cell. The KD value of the sample was calculated using Biacore Insight Evaluation Software.

Membrane proteome array

Membrane proteome array (MPA) screening were conducted at Integral Molecular, Inc (PA, USA). The MPA is a protein library composed of 6000 distinct human membrane protein clones, each overexpressed in live cells from expression plasmids. Each clone was individually transfected in separate wells of a 384-well plate followed by a 24 h incubation [42]. Cells expressing each individual MPA protein clone were arrayed in duplicate in a matrix format for high-throughput screening. Before screening on the MPA, the MUC18 scFv-Fc recombinant protein concentration for screening was determined on cells expressing positive (membrane-tethered Protein A) and negative (mock-transfected) binding controls, followed by detection by flow cytometry using a fluorescently-labeled secondary antibody. The MUC18 scFv-Fc recombinant protein was added to the MPA at the pre-determined concentration, and binding across the protein library was measured on an Intellicyt iQue using a fluorescently labeled secondary antibody. Each array plate contains both positive (Fc-binding) and negative (empty vector) controls to ensure plate-by-plate reproducibility. The MUC18 scFv-Fc recombinant protein interactions with any targets identified by MPA screening were confirmed in a second flow cytometry experiment using serial dilutions of the test antibody, and the target identity was re-verified by sequencing.

Plasmid construction and lentivirus production

The second-generation CAR molecules targeting MUC18 was designed to comprise of the signal peptide from colony stimulating factor 2 receptor alpha, scFv derived from pAb253, different hinge regions derived from IgG4, CD8α, or CD28, transmembrane domains from CD8α or CD28, co-stimulatory domain from 4-1BB or CD28, and intracellular wildtype (WT) domain or a mutant CD3ζ domain. The CAR genes were synthesized and inserted into the third-generation EF1α promoter-based lentiviral transfer plasmid pLenti6.3/V5 (ThermoFisher, Waltham, MA, USA). The CD3ζ domain was mutated by Fast Mutagenesis System (TransGen Biotech, China). Lentivirus stock was prepared by transient transfection of transfer plasmid, packaging plasmids (pLP1 and pLP2, ThermoFisher, Waltham, MA, USA) and envelope plasmid (pLP/VSVG, ThermoFisher, Waltham, MA, USA) to HEK293T cells using polyethyleneimine, collection of the culture medium 48 h and 72 h after transfection, ultrafiltration of the culture medium, and subsequent purification of the lentiviral particles using Core 700 chromatography (GE Healthcare, USA).

CAR-T cell production

Human T cells were activated by priming human CD3+ T cells isolated from peripheral blood mononuclear cells (PBMCs) of healthy donors with a CD3/CD28 Dynabeads (Thermo Fisher, USA) at 1.5:1 bead: cell ratio in X-VIVO medium (Lonza, Switzerland) with 500 U/mL hIL-2 (SL Pharm, China). One day after activation, T cells were transduced with MUC18-CAR encoding lentivirus and maintained at 1.5 × 106 cells/mL in T-cell medium containing X-VIVO media supplemented with 500 U/mL human interleukin-2 (IL-2). Beads were removed 4 days after activation. During the cell culture, CAR-T cell proliferation and viability were tested every 2 days with automated cell counter (Countess 3 FL, Thermo Fisher).

Primary melanoma cell isolation

Fresh tissues from melanoma patients stored in tissue storage solution (Miltenyi) were mechanically dissociated and digested in collagenase IV (200U/mL; ThermoFisher) and DNAse I (0.02% w/v; ThermoFisher). 4 volumes of stop solution (DPBS with 2mM EDTA and 2% FBS) was added to stop the digestion. Cell suspension was filtered by 70 μm cell filter and centrifuged at 300 g for 5 min. Precipitates were washed by stop solution and cultured by DMEM/F12 (ThermoFisher) with 10% FBS and 0.2% Primocin (Invivogen). Differential attachment method was used for purification of primary tumor cells. Briefly, cells were digested with 0.05% Trypsin after 72 h proliferation. Suspended cells were centrifuged and washed with PBS and incubated in polylysine coated dish for 30 min. Suspended cells were collected for further culture. Repeat this process for 2 to 3 times or until the cells become monomorphic colonization.

Flow cytometry

All flow cytometry measurements were performed on Northern Lights (Cytek Biosciences, USA). MUC18 expression on human melanoma cell lines (A375, GAK, HMV-II) and MUC18-knockout A375 cell line was detected by MUC18 scFv-hFc recombinant protein and PE anti-human IgG Fc Antibody (BioLegend, clone M1310G05). During the CAR-T cell culture, CD3 expression was detected by CD3 antibody (Biolegend, Cat#300439). CAR expression was detected by G4S linker rabbit mAb (Cell Signaling Technology, clone E7O2V, Alexa Fluor 488) every 2–3 days. For the CAR-T cell differentiation panel, CAR-T cells were stained with anti-human CD45RA antibody (BioLegend, clone HI100, BV510) and anti-human CD62L antibody (BioLegend, clone DREG-56, APC/Fire750). For the T cell exhaustion panel, T cells were stained with anti-human LAG-3 antibody (BioLegend, clone 11C3C65, PE/Cy7), PD-1 antibody (BioLegend, clone EH12.2H7, BV650).

Acute killing assays

Short-time cytotoxicity assay: CAR-T cells were cocultured with target cells at different E: T ratios in X-VIVO. The target cells A375 was stained with Calcein-AM (ThermoFisher, Cat#C3100MP) before coculture. After 5.5 h, supernatants were collected and analyzed with the Varioskan™ LUX multimode microplate reader (ThermoFisher, VL0000D0).

Real-time cytotoxicity assay: Target cells were seeded at 2 × 104/well in 96 well plates (E-plate96) for 24 h, followed by adding CAR-T cells at effector to target (E: T) ratios from 1:1 to 1:20. Cells were cultured in xCELLigence RTCA MP (Agilent Technologies) at 37℃, 5% CO2 and analyzed with RTCA Software Pro (Agilent Technologies).

Chronic antigen exposure (CAE) induced killing assay

MUC18 CAR-T cells were repeatedly stimulated with 1 × 105 A375-Luc+ cells every 2 days. Each round of restimulation started with replating fresh A375-Luc+ target cells, and adding CAR-T cells for co-culture one day later. About 1 × 105 of MUC18 CAR-T cells were initially added to the co-culture system, and tumor cell challenge was repeated every two days. Tumor rechallenge was terminated until a group of CAR-T cells could no longer kill the tumor cells. Then total amounts of tumor cells, T cells and CAR-T cells were analyzed by flow cytometry and calculated.

Cytokine measurements

To determine cytokine production by CAR-T cells, cell supernatant was harvested 24 h after co-culture with A375 cells. To measure cytokines levels, we used Cytokine Bead Array (CBA) Human Th1/Th2/Th17 CBA Kit (BD Biosciences; Cat# 560484) according to manufacturer protocol. In brief, capture beads (IL-2, TNF, IFN-γ, IL-4, IL-6, IL-10 and IL-17 A), PE-detection reagents and cell supernatant were co-incubated for 3 h. Beads were washed and analyzed by ACEA Flow Cytometer (ACEA Biosciences, NovoCyte 2060R).

Cell line-derived xenograft model

NCG mice (5–6 weeks old) were purchased from GemPharmatech (China) and maintained in pathogen free conditions. For A375 xenograft models, each mouse received a subcutaneous (s.c.) injection of 1 × 106 A375 cells. MUC18 CAR-T cells (2 × 106 or 3 × 106) were intravenously (i.v.) injected when tumor volumes reached ~ 100 mm3. Mice were evaluated by monitoring tumor volume and body weight every week. The level of CAR-T cells in peripheral blood was analyzed by ACEA Flow Cytometer. In the tumor rechallenging experiment, tumor-free mice after CAR-T treatment were rechallenged with 3 × 106 A375-Luc+ tumor cells for each mouse, and tumor relapse was monitored by weekly bioluminescence imaging on an IVIS spectrum instrument (PerkinElmer, Lumina II). Mice were euthanized at indicated days or at the endpoint of the experiment.

Patient-derived xenograft (PDX) model

Fresh tissues stored in tissue storage solution (Miltenyi) were cut into pieces with a diameter of 2 mm and subcutaneously inoculating into 6-week-old NCG female mice. After 30 days, mice with tumor volume reached approximately 250 mm3 were divided into five groups (6 in each group) and injected with 100 uL PBS, untransduced T cells (107 in total), or CAR-T cells (5 × 106 in total or 107 in total) via tail vein. Tumor volume was subsequently measured twice a week. When tumor volume reached larger than 2000 mm3, mice were sacrificed and samples of the liver, lung, brain, and tumor tissues were collected for subsequent hematoxylin and eosin (H&E) staining. All animal care and experimental procedures were carried out in accordance with Animal Care Ethics and were approved by the Medical Ethics Committee of the Beijing Cancer Hospital & Institute.

The migration and distribution of CAR-T cells in vivo

NCG mice (6–8 weeks old) were purchased from GemPharmatech (China) and maintained in pathogen free conditions. For 4H28T.28z CAR-T trafficking analysis, each mouse received a subcutaneous (s.c.) injection of 1.5 × 106 A375 cells. A total of 3 × 106 4H28T.28z CAR-T cells labeled with DiD dye (Beyotime, C1039) were intravenously injected when tumor volumes reached ~ 100 mm3. CAR-T cells were monitored by DiD dye fluorescence on an IVIS spectrum instrument (PerkinElmer, Lumina II) on day one, day four and day eight after CAR-T injection. Tumors, heart, liver, spleens, lungs, and kidneys were collected nine days after CAR-T infusion for ex vivo imaging.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software Inc., La Jolla, CA). For comparisons of three or more groups, a one- or two-way analysis of variance (ANOVA) was used followed by Tukey’s multiple comparisons test. For all experiments, P values of less than 0.05 were considered significant.

Data availability

All raw data and source data that support the findings of this study are available from the corresponding authors upon request.

Results

pAb253-derived ScFv bound to MUC18 that is highly expressed on melanoma cells

Previously we obtained a monoclonal antibody targeting MUC18 (pAb253) using a proteome-scale antibody array platform (PETAL). The pAb253-derived VH and VL domains were connected by a (GGGGS)3 linker and genetically fused to a human IgG Fc fragment, ultimately forming the single-chain variable fragment-human Fc (scFv-hFc) designated as YM03-253SV. Mucosal melanoma cell lines GAK and HMVII, as well as cutaneous melanoma cell line A375 were used to verify the binding of YM03-253SV to MUC18 by flow cytometry. Our results demonstrated that the YM03-253SV bound to MUC18 on all of the cell lines while MUC18-knockout in A375 cells abrogated its binding ability (Fig. 1A). Surface plasmon resonance (SPR) analysis showed that equilibrium KD for the scFv-hFc binding to MUC18 was 3.16 × 10− 8 M (Fig. 1B).

Fig. 1.

Fig. 1

Open in a new tab

YM03-253SV bound to MUC18 that was highly expressed on melanoma cells. (A) Flow cytometric analysis of MUC18 expression on melanoma cell lines using MUC18 scFv-hFc recombinant protein and PE anti-human IgG Fc Antibody. A375-MUC18-KO cells were generated via CRISPR-Cas9-mediated deletion in A375 cells. Control indicated A375 cells that were incubated with PE anti-human IgG Fc antibody only. (B) SPR analysis of the binding affinity between the MUC18 protein and the MUC18 scFv-hFc recombinant protein. The equilibrium dissociation constant KD of the MUC18 protein to the MUC18 scFv-hFc was 3.16 × 10− 8 M. The colored lines represent sensorgram curves obtained at different protein concentrations. The table below shows the affinity results of MUC18 scFv-hFc and human MUC18 protein. (C) Membrane Proteome Array results of MUC18 scFv-hFc binding specificity. The blue dot indicates that MUC18 scFv-hFc selectively bound to the human MUC18 protein

To identify the binding specificity of scFv-hFc, approximately 6,000 human cell membrane protein complementary DNAs were prepared and transfected into HEK293T cells, followed by flow cytometry analysis using the scFv-hFc (Fig. 1B). Screening the ~ 6,000 cell surface proteins revealed that the scFv-hFc specifically bound to MUC18-expressing cells (Fig. 1C), indicating minimal off-target binding by the scFv-hFc on non-target membrane proteins, eliminating the concern over nonspecific CAR-T toxicity for normal tissue.

MUC18 CAR-T containing the IgG4 hinge region and CD28 costimulatory signal domain demonstrates superior anti-tumor efficacy in melanoma models

To determine whether YM03-253SV can direct T cells to kill melanoma cells, the variable region of YM03-253SV was cloned into various CAR vectors, including 8 H/T.BBz1, 8 H/T.BBz2, 4H8T.BBz1, 28 H/T.28z, 28 H/T.28z3, and 4H28T.28z (Fig. 2A). The distance between T cells and target cells is crucial for the formation of an immunological synapse in T cells and subsequent killing of the target cells [43]. Therefore, YM03-253SV-CARs with decreasing hinge length (IgG4 hinge) were constructed to optimize the CAR structure (Fig. 2A). MUC18 CAR constructs were inserted into the lentiviral vector and subsequently transduced into human primary T cells. T cell transduction efficiency for all constructs ranged from 40 to 80% (Fig. 2B). Additionally, the growth curve analysis demonstrated that T cells expressing different CAR constructs exhibited similar rates of proliferation (Fig. S1A). Moreover, the proportions of stem cell memory T and naïve T (S&N) in the 4H8T.BBz1 group were slightly higher compared with other groups (Fig. S1B). Furthermore, the expression of exhaustion markers on the surface of the resting CAR-T cells was analyzed. It was observed that the CAR-T cells with 4-1BB signaling (8 H/T.BBz1, 8 H/T.BBz2 and 4H8T.BBz1) had lower proportion of exhaustion markers compared to those of CAR-T constructs bearing CD28 signaling domain (Fig. 2C).

Fig. 2.

Fig. 2

Open in a new tab

MUC18 CAR-T containing the IgG4 hinge region and CD28 costimulatory signal domain demonstrated superior anti-tumor efficacy in melanoma models. (A) Schema representation of the second-generation MUC18 CAR structures with distinct domains. (B) Representative flow cytometry analysis of MUC18-CAR transduction efficiency in primary T cells at different time points by detecting (G4S)3 linker expression. (C) Representative FACS plots (left) and quantification (right) of exhausted T cell subsets using PD-1 and LAG-3 as markers in CD3 + CAR-T cells. (D) In vitro cytotoxicity of MUC18 CAR-T cells against A375 cells, measured using Calcein-AM assay 5.5 h after co-culture (n = 3). (E) Cytolytic activity of MUC18 CAR-T cells against A375, GAK and HMVII cells in vitro. The cytolytic activity of different CAR-T cell groups was measured using RTCA, and the significance was tested at 16 h after co-culture (n = 3). (F) Experimental design of CAR-T cell dysfunction in an in vitro CAE stress test model, where MUC18 CAR-T cells were repeatedly stimulated with 1 × 105 A375 cells every 2 days. Arrows represent each round of restimulation with replating fresh A375 target cells the day before, and adding remaining CAR-T cells of the last round the day after. 1 × 105 CAR+ MUC18 CAR-T was added to the co-culture system at the first round of restimulation 11 days after T cell activation. FC, flow cytometry. (G) Total tumor burden remaining at the end of the CAE stress test quantified by using flow cytometry gated on GFP+ A375 target cells (n = 3). (H) Total T cells count of MUC18 CAR-Ts at the end of the CAE stress test, quantified using flow cytometry by gating on CD3+ T cells (n = 3). (I) Flowchart depicting the establishment of an A375 tumor xenograft model in NCG mice, followed by transfusion with UTD T cells or MUC18 CAR-T cells (n = 6 mice per group). (J) In vivo efficacy of MUC18 CAR-T cells in the A375 tumor mouse model. Tumor size was measured and monitored over time. (K) Quantification of MUC18 CAR-T cells (CD3+CAR+) in peripheral blood of mice, detected using flow cytometry over time (n = 6)

We then compared the functions of these CAR-T cells in vitro. MUC18 CAR-T cells were co-cultured with A375 cells for 5.5 h and CAR-Ts with CD28 co-stimulatory domain (28 H/T.28z, 28 H/T.28z3 and 4H28T.28z) had higher killing effects on A375 cells than other CAR-Ts (Fig. 2D). When cocultured with melanoma cell lines for 16 h, we obtained similar results except for the GAK cell line, in which 4H8T.BBz1 CAR-T cells exhibited comparable cytotoxicity to CAR-T cells with CD28 co-stimulatory domain (Fig. 2E). To determine whether various CAR-T constructs could prevent or delay CAR-T dysfunction, we used chronic antigen exposure (CAE) to conduct an in vitro stress test [44]. In this assay, MUC18 CAR-T cells were repeatedly stimulated with MUC18-expressing A375 cells, leading to dysfunction in the MUC18 CAR-T cells (Fig. 2F). We observed distinct proliferation and cytotoxicity profiles of different CAR-T constructs after repeated tumor-cell stimulation. After repeated stimulation, 4H8T.BBz1, 28 H/T.28z, 28 H/T.28z3 and 4H28T.28z displayed dramatically improved cytolytic activity resulting in reduced tumor burden compared with 8 H/T.BBz1 and 8 H/T.BBz2. Among the four CAR-Ts with the least remaining tumor burdens, 28 H/T.28z, 28 H/T.28z3 and 4H28T.28z cells exhibited more CAR-T cells (Fig. 2G, H).

To ascertain whether our in vitro findings correlated with in vivo efficacy, we evaluated the anti-tumor activity of MUC18 CAR-Ts with CD28 co-stimulatory domains as well as 4H8T.BBz1 CAR-T in the A375 model in NCG mice (Fig. 2I). MUC18 CAR-T cells eliminated the A375 tumors 18 days after injection while control untransduced (UTD) T cells had no effect on A375 tumor growth (Fig. 2J). These results demonstrated that our MUC18-CAR-Ts specifically eliminated MUC18-expressing cells in vivo. NCG mice treated with 4H28T.28z CAR-T cells exhibited faster tumor control compared with other CAR-T groups (Fig. 2J). CAR-T-cell proliferation was not always correlated with tumor control speed in vivo, as evidenced by 28 H/T.28z3 CAR-T cells with highest CAR+ T-cell number in peripheral blood and inferior anti-tumor ability compared with 4H28T.28z CAR-T cells (Fig. 2K). No obvious body weight reduction was observed among the groups throughout the experiment (Fig. S1C).

Collectively, these results demonstrate that human primary T cells expressing 4H28T.28z CAR with short IgG4 hinge and CD28 co-stimulatory domain potently kill MUC18-expressing A375 cells in vivo and thus chosen for further studies.

CAR molecule containing CD8 TMD in T cells demonstrate greater CAR expression and superior phenotype compared to CAR molecules with CD28 TMD

4H28T.28z-transduced CAR-T cells exhibited superior anti-tumor effect in vitro and in vivo, but resulted in relatively low virus titer and cell-surface CAR expression (Fig. 2B). To optimize lentiviral production efficiency and enhance surface CAR density, we rationally substituted the CD28 transmembrane domain (TMD) with its CD8α counterpart in the CAR architecture, designating this optimized construct as 4H8T.28z (Fig. 3A). Our prior studies revealed that CAR-T cells engineered with a modified CD3ζ intracellular domain containing four critical substitutions, including V2L, D9E, Q15K, and Y90F (designated z3), exhibited superior tumor suppression in murine xenograft models compared to their wild-type CD3ζ counterparts [45]. Building on these findings, we sought to determine whether analogous z3 domain incorporation could similarly enhance the therapeutic potential of MUC18-targeted CAR-T constructs (Fig. 3A). Both 4H8T.28z and 4H8T.28z3 CAR-T constructs yielded higher virus titers and surface CAR-expression levels compared with 4H28T.28z (Fig. 3B). Similar proliferation capacities were observed for 4H8T.28z and 4H8T.28z3 CAR-T cells compared with 4H28T.28z CAR-T cells (Fig. 3C). Moreover, 4H8T.28z and 4H8T.28z3 CAR-T cells had higher proportions of naïve T and stem cell memory T (S&N) cells compared with 4H28T.28z CAR-T cells (Fig. 3D-E). In terms of exhaustion levels, the ratio of exhaustion markers in the 4H8T.28z group was slightly lower than that in 4H28T.28z and 4H8T.28z3 groups (Fig. 3F-G).

Fig. 3.

Fig. 3

Open in a new tab

CAR molecule containing CD8 TMD in T cells demonstrated greater CAR expression and superior phenotype compared to CAR molecules with CD28 TMD. (A) Schematic representation of second-generation MUC18 CAR constructs. (B) Representative flow cytometry analysis (left) of positive MUC18 CAR-Ts at different time points as well as lentivirus titers (right) of various CAR-T constructs. (C) Representative proliferation curve of CAR-T cells in the absence of antigen. (D-E) Representative FACS plots (D) and percentage (E) of memory T cell subsets defined using CD45RA and CD62L as markers in CD3 CAR-T cells. (F-G) Representative FACS plots (F) and percentage (G) of exhaustion cell subsets using PD-1 and LAG-3 as markers in CD3 CAR. S&N: stem cell memory T cells and naïve T cells; CM: central memory T cells; EM: effector memory T cells; EFF: effector T cells

Taken together, 4H8T.28z CAR-T cells with CD8 TMD exhibit higher surface- CAR expression, more favorable differentiation phenotypes and lower exhaustion markers compared to the 4H28T.28z CAR-T cells.

CAR-T cells with CD8 transmembrane domain exhibit superior functionality in vitro and comparable antitumor effects in vivo compared with 4H28T.28z CAR-T cells

To further assess the effects of the CD8 TMD in CAR-T cells, we compared killing of 4H8T.28z and 4H8T.28z3 CAR-Ts with 4H28T.28z CAR-T in vitro. Using 1:5 or 1:10 E: T ratios against A375, we observed superior killing by 4H8T.28z CAR-T cells incorporating the CD8 TMD and WT CD3ζ signaling domain (Fig. 4A). In accordance with their enhanced cytolytic potential, 4H8T.28z CAR-T cells exhibited a distinct cytokine profile characterized by significant upregulation of pro-inflammatory mediators (IL-2, IFN-γ) coupled with marked downregulation of immunosuppressive cytokines (IL-4, IL-10) compared to 4H28T.28z counterparts (Fig. 4B). This Th1-skewed cytokine polarization pattern correlates with the observed cytotoxic efficacy, suggesting mechanistic linkage between cytokine milieu and antitumor functionality. Furthermore, chronic antigen exposure analysis revealed no significant difference in the killing ability of CAR-T cells among different groups (Fig. 4C). However, more 4H8T.28z and 4H8T.28z3 CAR-T cells remained in the co-culture system compared to the 4H28T.28z CAR-T cells (Fig. 4D, E).

Fig. 4.

Fig. 4

Open in a new tab

CAR-T cells with CD8 TMD exhibited superior functionality in vitro and comparable antitumor effects in vivo compared with 4H28T.28z CAR-T cells. (A) In vitro cytotoxicity by MUC18 CAR-T cells against A375 cells at different effector-to-target (E: T) ratios, the data from the co-culture at 24 h were subjected to statistical analysis. (B) Cytokine levels in the supernatant of MUC18 CAR-T cells co-cultured with A375 cells for 24 h at a E: T ratio of 1:2. (C) Total tumor burden remaining at the end of the CAE stress test quantified by using flow cytometry gated on GFP + A375 target cells (n = 3). (D) Total CD3+ CAR+ T cells and (E) CD3+ T of MUC18 CAR-Ts at the end of the CAE stress test quantified by using flow cytometry (n = 3). (F) Flowchart of establishing A375-Luc+ tumor xenograft model in NCG mice and transfusion with UTD T cells or MUC18 CAR-Ts (n = 6 mice per group). Rechallenge was performed with A375−Luc+ cells at 36 days after MUC18 CAR-T transfusion. (G) Summary of tumor volumes after treatment with MUC18 CAR-T cells (n = 6 tumors per group). (H) Flux (P/s) values of tumor burden after tumor rechallenge assessed by IVIS imaging (n = 6). (I) The number of CD3+ CAR+ MUC18 CAR-T cell in peripheral blood of mice detected by flow cytometry over time (n = 6). (J) Weight of treated mice as change from baseline over the course of the experiment (n = 6). *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 by repeated measure two-way ANOVA for multiple comparisons or ordinary one-way ANOVA

To further evaluate the efficacy of the above three MUC18 CAR-Ts in suppressing A375 xenografts in vivo, A375 cells were transplanted into NCG mice, followed by intravenous infusions of control UTD T cells or MUC18 CAR-T cells. We also investigated the effect of MUC18 CAR-T cells inhibiting tumor recurrence after A375 cell line rechallenge (Fig. 4F). All three MUC18 CAR-T cells eradicated tumor cells around day 22 after CAR-T cell injection and kept mice tumor-free until day 36 when MUC18 CAR-T-treated mice were rechallenged with A375 expressing firefly luciferase (A375-luc+). Though not significant, A375-Luc+ tumors appeared to be inhibited faster in the 4H8T.28z group compared to the other groups (Fig. 4G, H). Additionally, we dynamically monitored the proliferation of CAR-T cells in the peripheral blood and observed that MUC18 CAR-T cells showed similar proliferation profiles (Fig. 4I). Additionally, there was no statistically significant variation in body weight observed among the various groups of CAR-T cells in mice (Fig. 4J). To investigate the in vivo biodistribution of 4H8T.28z CAR-T cells, tumor-bearing mice was administered DiD-labeled CAR-T cells or saline via intravenous injection. Longitudinal tracking revealed distinct temporal dynamics: CAR-T cells initially localized predominantly in the liver (24 h post-infusion), followed by progressive redistribution to tumor sites with residual hepatic accumulation observed at day 4 post-administration (Fig. S2A, B). Quantitative analysis of tissue distribution demonstrated significantly elevated CAR-T cell signals in tumor tissues and reticuloendothelial organs (liver, spleen, and lungs) compared to other organs (heart and kidneys) (Fig. S2B, C). This organotropic distribution pattern suggests preferential homing to both neoplastic lesions and immune-filtering organs. These data suggest that 4H8T.28z and 4H8T.28z3 CAR-Ts with CD8 TMD significantly enhance surface CAR expression and in vitro functionality while exhibiting comparable antitumor effects in vivo against melanoma.

The MUC18 CAR-T treatment effectively inhibits tumor growth in a mucosal melanoma PDX model

To evaluate the cytotoxicity and safety of MUC18 CAR-T, a primary mucosal melanoma cell line was isolated from fresh tissue. Real-time cytotoxicity assay indicated that 4H8T.28z CAR-T cells could effectively inhibit the growth of MUC18-expressing primary melanoma cells in a dose-dependent manner. More than 90% of target cells were killed by 4H8T.28z CAR-T after 70 h of coculture. Instead, UTD T cells had no cytotoxic effect (Fig. 5A-D). In vivo assay was carried out as described in Fig. 5E, 4H8T.28z and 4H8T.28z3 CAR-T cells were injected into primary mucosal melanoma tumor-bearing mice. After 30 days of treatment, all CAR-T groups received rapid tumor regression (Fig. 5F and Fig S3A). CAR-T cells proliferated rapidly after injection without any significant difference among various CAR-T groups (Fig. 5G, H). Despite potent anti-tumor effects (Fig. 5I), animals in all CAR-T groups exhibited no weight loss or clinical signs of toxicity. Hematoxylin-eosin staining indicated the histomorphological structures of major organs were intact, including liver and lung which have a high rate of CAR-T infiltration in A375 xenografts model (Fig. 5J). We conclude that 4H8T.28z and 4H8T.28z3 CAR-T cells could effectively control tumors in the PDX model without evidence of toxicity.

Fig. 5.

Fig. 5

Open in a new tab

The MUC18 CAR-T treatment effectively inhibited tumor growth in a mucosal melanoma PDX model. (A-D) In vitro cytotoxicity assay of primary melanoma cells, primary melanoma cells were cocultured with UTD or 4H8T.28z3 CAR-T cells at ratios of 1:1, 1:5, 1:10, and 1:20. (E) Schematic representation of PDX model. (F) In vivo efficacy evaluation in a melanoma PDX mouse model. Tumor size was measured and monitored over time (n = 6 mice per group). (G) Number of CD3+ T cells and (H) MUC18 CD3+ CAR+ CAR-T cells in peripheral blood of mice detected by flow cytometry over time. (I) Weights of treated mice over the course of the experiment. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 by repeated measure two-way ANOVA for multiple comparisons or ordinary one-way ANOVA. (J) Representative H&E staining of the liver, lungs, cerebrum, and cerebellum from treated mice

Discussion

Acral, mucosal, and uveal melanomas constitute rare melanoma subtypes with low incidence rates yet harbor dismal prognoses due to poor responsiveness to standard melanoma therapies [46]. Furthermore, the lower frequency of BRAF/NRAS mutations (lower than 20%) in mucosal melanoma limits the application of MAPK pathway inhibitors [5]. Immunotherapies exhibit variable responses and are hampered by the absence of highly sensitive predictive biomarkers. Comparative analyses have revealed the median progression-free survival after immunotherapy in Chinese melanoma patients is 3.6 months, with an objective response rate (ORR) of 18.6%, significantly lower than that observed in Western populations (mPFS = 6.3 months; ORR = 39.5%) [10]. The therapeutic landscape for advanced melanoma refractory to first-line immune checkpoint inhibitors and second-line targeted therapies has entered an era of biological innovation, with cellular therapies emerging as particularly promising modalities. TIL (Tumor-Infiltrating Lymphocyte) therapy has established the most robust efficacy profile, with phase III data showing doubling of objective response rates compared to ipilimumab (49% vs. 21%) [12]. However, the logistical constraints of requiring fresh tumor specimens (available in only 60–70% of cases in registry studies) and four-to-six-week manufacturing times remain significant barriers to widespread adoption [47].

CAR-T strategies face unique challenges in melanoma compared to hematological malignancies. A phase 1 trial of GD2-specific CAR-T cells in patients with GD2-positive metastatic melanoma indicated robust efficacy. Of 12 patients treated with combined BRAF-MEK inhibition and GD2 CAR-T, 5 (42%) achieved partial responses according to Response Evaluation Criteria in Solid Tumors version 1.1 [48]. Emerging CAR-T cell therapies include CAR constructs against TYRP1 showing antitumor activity in preclinical models [49], though clinical validation is pending. Notably, approximately 60–90% of mucosal, acral, and cutaneous subtype melanoma patients exhibit MUC18 overexpression, positioning MUC18 CAR-T therapy as a promising intervention for this molecularly defined patient subgroup [41]. The elevated expression of MUC18 in both tumor cells and the associated vasculature also suggests a therapeutic opportunity for dual-targeting with MUC18-specific CAR-T cells [41]. As the limited infiltration of CAR-T cells within the complex microenvironment of solid tumors is a key barrier in the development of effective CAR-T therapies for solid malignancies, this dual-targeting strategy may enhance CAR-T cell infiltration by diminishing the physical barriers within the tumor microenvironment. The dual-targeting capability of MUC18-directed CAR-T cells also enabling combinatorial treatment approaches. For example, MUC18-targeted CAR-T cells could potentially enhance the tumor penetration of chemotherapeutic agents or targeted therapies, thereby amplifying their synergistic therapeutic effects. Additionally, the combination of MUC18-targeted CAR-T cells with anti-angiogenic therapies may further facilitate CAR-T cell infiltration into the tumor, optimizing the efficacy of immunotherapeutic interventions.

Our study aims to engineer MUC18-targeting CAR-T cells to target mucosal melanoma. Firstly, we screened an ideal scFv targeting MUC18. Subsequently, we constructed various second-generation CAR constructs and assessed cytotoxicity of MUC18 CAR-T cells in vitro and in vivo. These experiments demonstrate that CAR constructs containing the IgG4 hinge and CD28 costimulatory domain (4H28T.28z) has better functions both in vitro and in vivo. However, CAR expression and the virus titer are relatively low for 4H28T.28z CAR-T compared with other constructs. Thus, we substitute the CD28 TMD with CD8 TMD to see if CD8 TMD could improve the CAR expression. Surprisingly, CD8 TMD substitution enhances surface CAR expression. Additionally, MUC18 CAR-T cells with CD8 TMD and CD28 co-stimulatory domain exhibit superior anti-tumor efficacies on cutaneous melanoma CDX and mucosal melanoma PDX models. To our knowledge, this study constitutes the first pre-clinical validation of MUC18 CAR-T therapy in melanoma, with particular efficacy demonstrated in mucosal melanoma. Overall, these results demonstrate the potential effectiveness of MUC18 CAR-T to benefit patients with cutaneous and mucosal melanoma.

Conclusions

In this study, a series of MUC18 targeted CAR-T was designed and tested in mucosal melanoma, one of the candidates with gG4-derived hinge domain and CD28 co-stimulatory domain exhibited best antitumor efficiency. This study suggests the potential of MUC18 as a target for CAR-T therapy in mucosal melanoma, and presents a strategy for optimizing the structure of CAR-T therapy for solid tumors.

Although our research on advanced melanoma has yielded promising results, there are still challenges associated with implementing MUC18-targeted CAR-T cell therapy. (1) Optimize CAR structures and functions to enhance T cell proliferation, targeting, and persistence. (2) Exploring potential combination therapy of immunotherapy and targeted therapy to overcome immunosuppressive immune microenvironment and provide potential therapeutic options for refractory melanoma. (3) Extensive studies of MUC18 in microenvironment are required, especially the potential influence of targeting MUC18+ cell in tumor microenvironment. Of note, MUC18 was also known as a tumor vascular marker, elimination of MUC18-positive cells in some tumor tissue may alter the angiogenesis and influence the immune infiltration. Besides, targeting MUC18-positive tumor vascular cells may result in a synergetic anti-tumor effect in MUC18-CAR-T treatment, and provide a potential target for tumors which do not express MUC18 as a marker in its own.

Electronic supplementary material

Below is the link to the electronic supplementary material.

12967_2025_6365_MOESM1_ESM.tiff (146.9KB, tiff)

Fig S1. MUC18 CAR-T containing the IgG4 hinge region and CD28 costimulatory signal domain demonstrates superior anti-tumor efficacy in melanoma models. (A) Representative proliferation of MUC18 CAR-T cells in the absence of antigen. UTD, untransduced T cells. (B) Representative FACS plots (left) and percentage (right) of memory cell subsets defined using CD45RA and CD62L as markers in CD3 CAR. S&N: stem cell memory T cells and naïve T cells; CM: central memory T cells; EM: effector memory T cells; EFF: effector T cells. (C) Mouse body weight was measured over time (n = 6). *p ≤ 0.05, by repeated measure two-way ANOVA for multiple comparisons or ordinary one-way ANOVA

12967_2025_6365_MOESM2_ESM.tiff (242.8KB, tiff)

Fig S2. CAR-T cells with CD8 TMD exhibit superior functionality compared to CAR-T cells with CD28 TMD both in vitro and in vivo. (A) Representative fluorescent image of whole-body distribution of DiD-labeled 4H8T.28z CAR-T cells at indicated time points and (B) at Day 9. (C) Representative fluorescent image and quantitation of DiD-labeled 4H8T.28z CAR-T cells accumulation in different tissues nine days after infusion

Acknowledgements

We would like to thank Beijing Imunopharm Technology Corporation for technique supporting and Multitude Therapeutics Corporation for English language editing.

Abbreviations

MUC18

Mucin 18

CAR-T

Chimeric antigen receptor-T

scFv

Single chain antibody fragment

PDX

Patient-derived xenograft

ACT

Adoptive cell immunotherapy

TIL

Tumor-infiltrating lymphocyte

FBS

Fetal bovine serum

STR

Short tandem repeat

sgRNA

Single guide RNA

SPR

Surface plasmon resonance

MPA

Membrane proteome array

WT

Wildtype

PBMCs

Peripheral blood mononuclear cells

IL-2

Interleukin-2

CAE

Chronic antigen exposure

CBA

Cytokine Bead Array

PETAL

Proteome-scale antibody array platform

UTD

Untransduced

TMD

Transmembrane domain

Author contributions

Conceptualization, Y. K., Y.D. and S.L.; Formal Analysis, F.Z., H.D. and K.L.; Investigation, F.Z., M.L., H.D., K.L. and Q.G.; Writing– Original Draft, F.Z., S.F., H.D. and K.L.; Writing– Review & Editing, J.S., T.H., X.L., Y.T., L.W., Q.L. and X.M.; Supervision, Y. K., Y.D. and S.L.; Project Administration, Y. K., Y.D. and S.L.; Funding Acquisition, Y. K., Y.D. and S.L.

Funding

This study was supported by National Key R&D Program of China (2019YFA0904404), Natural Science Foundation of China (82272848), National Key R&D Program of China, Stem Cell Research and Organ Repair (2022YFA1106500) and Beijing Engineering Research Center of Research and Development of New Antitumor Drugs and New Technologies.

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

All animal care and experimental procedures were carried out in accordance with Animal Care Ethics and were approved by the Medical Ethics Committee of the Beijing Cancer Hospital & Institute.

Consent for publication

Not applicable.

Conflicts of interest

The authors declare no potential conflicts of interest.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Fenghao Zhang, Haizhen Du and Kaiping Liu contributed equally to this work.

Contributor Information

Shu-Hui Liu, Email: shuhui.liu@multitudetherapeutics.com.

Yanping Ding, Email: dingyanping@imunopharm.com.

Yan Kong, Email: k-yan08@163.com.

References

  • 1.Nassar KW, Tan AC. The mutational landscape of mucosal melanoma. Semin Cancer Biol. 2020;61:139–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lian B, Si L, Cui C, Chi Z, Sheng X, Mao L, et al. Phase II randomized trial comparing high-dose IFN-α2b with Temozolomide plus cisplatin as systemic adjuvant therapy for resected mucosal melanoma. Clin Cancer Res. 2013;19(16):4488–98. [DOI] [PubMed] [Google Scholar]
  • 3.McLaughlin CC, Wu XC, Jemal A, Martin HJ, Roche LM, Chen VW. Incidence of noncutaneous melanomas in the U.S. Cancer. 2005;103(5):1000–7. [DOI] [PubMed] [Google Scholar]
  • 4.Chen W, Zheng R, Baade PD, Zhang S, Zeng H, Bray F, et al. Cancer statistics in China, 2015. CA Cancer J Clin. 2016;66(2):115–32. [DOI] [PubMed] [Google Scholar]
  • 5.Hayward NK, Wilmott JS, Waddell N, Johansson PA, Field MA, Nones K, et al. Whole-genome landscapes of major melanoma subtypes. Nature. 2017;545(7653):175–80. [DOI] [PubMed] [Google Scholar]
  • 6.Newell F, Kong Y, Wilmott JS, Johansson PA, Ferguson PM, Cui C, et al. Whole-genome landscape of mucosal melanoma reveals diverse drivers and therapeutic targets. Nat Commun. 2019;10(1):3163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chang AE, Karnell LH, Menck HR. The National cancer data base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American college of surgeons commission on cancer and the American cancer society. Cancer. 1998;83(8):1664–78. [DOI] [PubMed] [Google Scholar]
  • 8.Si L, Zhang X, Shu Y, Pan H, Wu D, Liu J, et al. A phase Ib study of pembrolizumab as Second-Line therapy for Chinese patients with advanced or metastatic melanoma (KEYNOTE-151). Transl Oncol. 2019;12(6):828–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Tang B, Chi Z, Chen Y, Liu X, Wu D, Chen J, et al. Safety, efficacy, and biomarker analysis of Toripalimab in previously treated advanced melanoma: results of the POLARIS-01 multicenter phase II trial. Clin Cancer Res. 2020;26(16):4250–9. [DOI] [PubMed] [Google Scholar]
  • 10.Bai X, Shoushtari AN, Betof Warner A, Si L, Tang B, Cui C, et al. Benefit and toxicity of programmed death-1 Blockade vary by ethnicity in patients with advanced melanoma: an international multicentre observational study. Br J Dermatol. 2022;187(3):401–10. [DOI] [PubMed] [Google Scholar]
  • 11.Depil S, Duchateau P, Grupp SA, Mufti G, Poirot L. Off-the-shelf’ allogeneic CAR T cells: development and challenges. Nat Rev Drug Discov. 2020;19(3):185–99. [DOI] [PubMed] [Google Scholar]
  • 12.Rohaan MW, van den Borch TH, Met Ö, Kessels R, Geukes Foppen MH, et al. Tumor-Infiltrating lymphocyte therapy or ipilimumab in advanced melanoma. N Engl J Med. 2022;387(23):2113–25. [DOI] [PubMed] [Google Scholar]
  • 13.Munshi NC, Anderson LD Jr., Shah N, Madduri D, Berdeja J, Lonial S, et al. Idecabtagene vicleucel in relapsed and refractory multiple myeloma. N Engl J Med. 2021;384(8):705–16. [DOI] [PubMed] [Google Scholar]
  • 14.Martin T, Usmani SZ, Berdeja JG, Agha M, Cohen AD, Hari P, et al. Ciltacabtagene autoleucel, an Anti-B-cell maturation antigen chimeric antigen receptor T-Cell therapy, for relapsed/refractory multiple myeloma: CARTITUDE-1 2-Year Follow-Up. J Clin Oncol. 2023;41(6):1265–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Shi D, Shi Y, Kaseb AO, Qi X, Zhang Y, Chi J, et al. Chimeric antigen Receptor-Glypican-3 T-Cell therapy for advanced hepatocellular carcinoma: results of phase I trials. Clin Cancer Res. 2020;26(15):3979–89. [DOI] [PubMed] [Google Scholar]
  • 16.Qi C, Gong J, Li J, Liu D, Qin Y, Ge S, et al. Claudin18.2-specific CAR T cells in Gastrointestinal cancers: phase 1 trial interim results. Nat Med. 2022;28(6):1189–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Liu H, Xu Y, Xiang J, Long L, Green S, Yang Z, et al. Targeting Alpha-Fetoprotein (AFP)-MHC complex with CAR T-Cell therapy for liver cancer. Clin Cancer Res. 2017;23(2):478–88. [DOI] [PubMed] [Google Scholar]
  • 18.Chinnasamy D, Yu Z, Theoret MR, Zhao Y, Shrimali RK, Morgan RA, et al. Gene therapy using genetically modified lymphocytes targeting VEGFR-2 inhibits the growth of vascularized syngenic tumors in mice. J Clin Invest. 2010;120(11):3953–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Inoo K, Inagaki R, Fujiwara K, Sasawatari S, Kamigaki T, Nakagawa S, et al. Immunological quality and performance of tumor vessel-targeting CAR-T cells prepared by mRNA-EP for clinical research. Mol Ther Oncolytics. 2016;3:16024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Chinnasamy D, Tran E, Yu Z, Morgan RA, Restifo NP, Rosenberg SA. Simultaneous targeting of tumor antigens and the tumor vasculature using T lymphocyte transfer synergize to induce regression of established tumors in mice. Cancer Res. 2013;73(11):3371–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Gargett T, Yu W, Dotti G, Yvon ES, Christo SN, Hayball JD, et al. GD2-specific CAR T cells undergo potent activation and deletion following antigen encounter but can be protected from activation-induced cell death by PD-1 Blockade. Mol Ther. 2016;24(6):1135–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yvon E, Del Vecchio M, Savoldo B, Hoyos V, Dutour A, Anichini A, et al. Immunotherapy of metastatic melanoma using genetically engineered GD2-specific T cells. Clin Cancer Res. 2009;15(18):5852–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Yang M, Tang X, Zhang Z, Gu L, Wei H, Zhao S, et al. Tandem CAR-T cells targeting CD70 and B7-H3 exhibit potent preclinical activity against multiple solid tumors. Theranostics. 2020;10(17):7622–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Simon B, Harrer DC, Schuler-Thurner B, Schuler G, Uslu U. Arming T cells with a gp100-Specific TCR and a CSPG4-Specific CAR using combined DNA- and RNA-Based receptor transfer. Cancers (Basel). 2019;11(5). [DOI] [PMC free article] [PubMed]
  • 25.Robbins PF, Morgan RA, Feldman SA, Yang JC, Sherry RM, Dudley ME, et al. Tumor regression in patients with metastatic synovial cell sarcoma and melanoma using genetically engineered lymphocytes reactive with NY-ESO-1. J Clin Oncol. 2011;29(7):917–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Schmidt P, Kopecky C, Hombach A, Zigrino P, Mauch C, Abken H. Eradication of melanomas by targeted elimination of a minor subset of tumor cells. Proc Natl Acad Sci U S A. 2011;108(6):2474–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Zhang Z, Jiang C, Liu Z, Yang M, Tang X, Wang Y, et al. B7-H3-Targeted CAR-T cells exhibit potent antitumor effects on hematologic and solid tumors. Mol Ther Oncolytics. 2020;17:180–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Soltantoyeh T, Akbari B, Karimi A, Mahmoodi Chalbatani G, Ghahri-Saremi N, Hadjati J et al. Chimeric antigen receptor (CAR) T cell therapy for metastatic melanoma: challenges and road ahead. Cells. 2021;10(6). [DOI] [PMC free article] [PubMed]
  • 29.Lehmann JM, Holzmann B, Breitbart EW, Schmiegelow P, Riethmüller G, Johnson JP. Discrimination between benign and malignant cells of melanocytic lineage by two novel antigens, a glycoprotein with a molecular weight of 113,000 and a protein with a molecular weight of 76,000. Cancer Res. 1987;47(3):841–5. [PubMed] [Google Scholar]
  • 30.Lehmann JM, Riethmüller G, Johnson JP. MUC18, a marker of tumor progression in human melanoma, shows sequence similarity to the neural cell adhesion molecules of the Immunoglobulin superfamily. Proc Natl Acad Sci U S A. 1989;86(24):9891–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Shih IM. The role of CD146 (Mel-CAM) in biology and pathology. J Pathol. 1999;189(1):4–11. [DOI] [PubMed] [Google Scholar]
  • 32.Sumardika IW, Youyi C, Kondo E, Inoue Y, Ruma IMW, Murata H, et al. β-1,3-Galactosyl-O-Glycosyl-Glycoprotein β-1,6-N-Acetylglucosaminyltransferase 3 increases MCAM stability, which enhances S100A8/A9-Mediated cancer motility. Oncol Res. 2018;26(3):431–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Johnson JP, Rummel MM, Rothbächer U, Sers C. MUC18: A cell adhesion molecule with a potential role in tumor growth and tumor cell dissemination. Curr Top Microbiol Immunol. 1996;213(Pt 1):95–105. [DOI] [PubMed] [Google Scholar]
  • 34.Mohammadi M, Nejatollahi F, Ghasemi Y, Faraji SN. Anti-Metastatic and Anti-Invasion effects of a specific Anti-MUC18 ScFv antibody on breast cancer cells. Appl Biochem Biotechnol. 2017;181(1):379–90. [DOI] [PubMed] [Google Scholar]
  • 35.Feng R, Wang Y, Ramachandran V, Ma Q, May MM, Li M, et al. Characterization of novel neutralizing mouse monoclonal antibody JM1-24-3 developed against MUC18 in metastatic melanoma. J Exp Clin Cancer Res. 2020;39(1):273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Zhang R, Chen X, Chen S, Tang J, Chen F, Lin Y, et al. Inhibition of CD146 lessens uveal melanoma progression through reducing angiogenesis and vasculogenic mimicry. Cell Oncol (Dordr). 2022;45(4):557–72. [DOI] [PubMed] [Google Scholar]
  • 37.Yan X, Lin Y, Yang D, Shen Y, Yuan M, Zhang Z, et al. A novel anti-CD146 monoclonal antibody, AA98, inhibits angiogenesis and tumor growth. Blood. 2003;102(1):184–91. [DOI] [PubMed] [Google Scholar]
  • 38.Stalin J, Nollet M, Garigue P, Fernandez S, Vivancos L, Essaadi A, et al. Targeting soluble CD146 with a neutralizing antibody inhibits vascularization, growth and survival of CD146-positive tumors. Oncogene. 2016;35(42):5489–500. [DOI] [PubMed] [Google Scholar]
  • 39.Liu J, Kang L, Smith S, Wang C. Transmembrane MUC18 targeted polydopamine nanoparticles and a mild photothermal effect synergistically disrupt actin cytoskeleton and migration of cancer cells. Nano Lett. 2021;21(22):9609–18. [DOI] [PubMed] [Google Scholar]
  • 40.Yu X, Liu J, Bauer A, Wei X, Smith S, Ning S, et al. Enhancing tumor endothelial permeability using MUC18-targeted gold nanorods and mild hyperthermia. J Colloid Interface Sci. 2024;676:101–9. [DOI] [PubMed] [Google Scholar]
  • 41.Shi J, Jiao T, Guo Q, Weng W, Ma L, Zhang Q, et al. A cell Surface-Binding antibody atlas nominates a MUC18-Directed antibody-Drug conjugate for targeting melanoma. Cancer Res. 2023;83(22):3783–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Tucker DF, Sullivan JT, Mattia KA, Fisher CR, Barnes T, Mabila MN, et al. Isolation of state-dependent monoclonal antibodies against the 12-transmembrane domain glucose transporter 4 using virus-like particles. Proc Natl Acad Sci U S A. 2018;115(22):E4990–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Liu D, Badeti S, Dotti G, Jiang JG, Wang H, Dermody J, et al. The role of immunological synapse in predicting the efficacy of chimeric antigen receptor (CAR) immunotherapy. Cell Commun Signal. 2020;18(1):134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Good CR, Aznar MA, Kuramitsu S, Samareh P, Agarwal S, Donahue G, et al. An NK-like CAR T cell transition in CAR T cell dysfunction. Cell. 2021;184(25):6081–e10026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ding YLX, He T, Liang M, Qi F. Beijing Imunopharm Technology Co., Ltd, CN116836282B.
  • 46.Dharanipragada P, Zhang X, Liu S, Lomeli SH, Hong A, Wang Y, et al. Blocking genomic instability prevents acquired resistance to MAPK inhibitor therapy in melanoma. Cancer Discov. 2023;13(4):880–909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Yu Z, Shi J, Fang Y, Zhao Y, Xu A, Li N. Developing innovative strategies of tumor–infiltrating lymphocyte therapy for tumor treatment. Oncol Rep. 2024;51(6). [DOI] [PMC free article] [PubMed]
  • 48.Gargett T, Truong NTH, Gardam B, Yu W, Ebert LM, Johnson A et al. Safety and biological outcomes following a phase 1 trial of GD2-specific CAR-T cells in patients with GD2-positive metastatic melanoma and other solid cancers. J Immunother Cancer. 2024;12(5). [DOI] [PMC free article] [PubMed]
  • 49.Jilani S, Saco JD, Mugarza E, Pujol-Morcillo A, Chokry J, Ng C, et al. CAR-T cell therapy targeting surface expression of TYRP1 to treat cutaneous and rare melanoma subtypes. Nat Commun. 2024;15(1):1244. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

12967_2025_6365_MOESM1_ESM.tiff (146.9KB, tiff)

Fig S1. MUC18 CAR-T containing the IgG4 hinge region and CD28 costimulatory signal domain demonstrates superior anti-tumor efficacy in melanoma models. (A) Representative proliferation of MUC18 CAR-T cells in the absence of antigen. UTD, untransduced T cells. (B) Representative FACS plots (left) and percentage (right) of memory cell subsets defined using CD45RA and CD62L as markers in CD3 CAR. S&N: stem cell memory T cells and naïve T cells; CM: central memory T cells; EM: effector memory T cells; EFF: effector T cells. (C) Mouse body weight was measured over time (n = 6). *p ≤ 0.05, by repeated measure two-way ANOVA for multiple comparisons or ordinary one-way ANOVA

12967_2025_6365_MOESM2_ESM.tiff (242.8KB, tiff)

Fig S2. CAR-T cells with CD8 TMD exhibit superior functionality compared to CAR-T cells with CD28 TMD both in vitro and in vivo. (A) Representative fluorescent image of whole-body distribution of DiD-labeled 4H8T.28z CAR-T cells at indicated time points and (B) at Day 9. (C) Representative fluorescent image and quantitation of DiD-labeled 4H8T.28z CAR-T cells accumulation in different tissues nine days after infusion

Data Availability Statement

All raw data and source data that support the findings of this study are available from the corresponding authors upon request.

Not applicable.

Articles from Journal of Translational Medicine are provided here courtesy of BMC

RESOURCES