Abstract
Chimeric antigen receptor (CAR)-T therapy has transformed management of hematological malignancies, yet efficacy in solid tumors is limited by scarcity of tumor-restricted antigens. Here we characterize SLC35D3, a nucleotide-sugar transporter that we previously identified through integrated transcriptomic and protein-level analyses, as a novel therapeutic target and provide the first preclinical evaluation of SLC35D3-targeted CAR-T cells. We generated rat monoclonal antibodies against human SLC35D3, selected clone #021, humanized it to produce h021-12, and incorporated its single-chain variable fragment into a second-generation CD8α/4-1BB/CD3ζ CAR. For functional assays, we established isogenic tumor-cell models—knockout (KO), wild-type (WT), and overexpressing (OE)—and confirmed graded SLC35D3 surface expression by flow cytometry. In vitro functional assays revealed that h021-12 CAR-T cells exhibited antigen-density-dependent cytotoxicity and interferon-γ release. In tumor engraftment-inhibition xenograft studies, a single intravenous dose of h021-12 CAR-T cells profoundly suppressed OE tumor growth while sparing KO tumors and causing no discernible body weight loss, indicating favorable safety and strict antigen dependence. Activity against WT tumors with physiological SLC35D3 levels remained modest, highlighting antigen-density dependence and avenues for improvement. Overall, our findings provide compelling preclinical proof of concept for targeting SLC35D3 and underscore the need for further optimization to enable future translation into clinical studies in precision oncology.
Keywords: SLC35D3, CAR-T, Antibody generation
Highlights
-
•
First functional validation of SLC35D3 as a target for CAR-T cell therapy.
-
•
Novel anti-SLC35D3 antibodies show high specificity and cross-reactivity.
-
•
h021-12 CAR-T cells exhibit antigen-dependent cytotoxicity and cytokine release.
-
•
Single CAR-T dose profoundly suppresses high-antigen tumors without toxicity.
1. Introduction
Although CAR T-cell therapy has revolutionized the treatment of hematologic malignancies [1,2], extending CAR-T efficacy to solid tumors remains a major challenge. Tumor heterogeneity, the immunosuppressive tumor microenvironment (TME), and the scarcity of antigens that are abundantly expressed on cancer cells but silent in vital normal tissues collectively limit therapeutic windows [[3], [4], [5]]. Consequently, identifying and validating novel tumor-selective surface antigens is essential for the rational design of next-generation CAR-T cells.
Recent transcriptomic and proteomic analyses have facilitated systematic discovery of candidate targets for antibody-based and cell-based immunotherapies [6,7]. Our previous study employed the Genotype-Tissue Expression (GTEx) and The Cancer Genome Atlas (TCGA) datasets to identify SLC35D3 (solute carrier family 35, member D3)—a little-studied nucleotide-sugar transporter. Our companion study further demonstrated that SLC35D3 expression is preferentially detected in selected tumor types, whereas protein expression in major normal organs such as brain, heart, lung, liver, and kidney was minimal or undetectable, with weak staining mainly restricted to scattered endocrine/neuroendocrine cell populations [8]. However, no antibody-based reagents or cellular immunotherapies targeting SLC35D3 had been reported.
To address this gap, we generated a humanized anti-SLC35D3 monoclonal antibody (h021-12) and incorporated its single-chain variable fragment (scFv) into a second-generation CAR [9] containing a CD8α hinge and transmembrane domain, a 4-1BB co-stimulatory motif, and a CD3ζ signaling domain (hereinafter referred to as “h021-12 CAR”). Using isogenic tumor-cell models with varying SLC35D3 expression (KO, WT, and OE), we evaluated h021-12 CAR-T cell function through in vitro and in vivo studies. This study establishes the first proof of concept for SLC35D3-targeted CAR-T therapy in solid tumors.
2. Materials and methods
2.1. Cell lines
Human pancreatic islet carcinoma QGP-1, colorectal carcinoma HCT 116, and gastric carcinoma SNU-16 cell lines were obtained from ATCC (USA) or JCRB (Japan). QGP-1 and SNU-16 cells endogenously express SLC35D3, whereas HCT 116 cells are SLC35D3-negative. To enable live-cell imaging, QGP-1 and HCT 116 cells were transduced with a lentiviral vector encoding green fluorescent protein (GFP), generating GFP-labelled parental derivatives designated QGP-1 WT and HCT 116 Neg, respectively. For functional studies, isogenic derivatives were established: QGP-1 OE stably overexpressing SLC35D3-GFP via lentiviral transduction; QGP-1 KO and SNU-16 KO generated by CRISPR-Cas9-mediated disruption of the SLC35D3 gene. All cell lines were maintained in RPMI-1640 supplemented with 10% heat-inactivated fetal bovine serum (FBS; incubated at 56°C for 30 min) at 37°C in a humidified 5% CO2 incubator, with puromycin selection where appropriate.
2.2. Antibody generation and profiling
Chimeric monoclonal antibodies (#021 and #114) against SLC35D3 were generated as previously described [10], by rat DNA immunization and hybridoma screening, and cloned into human IgG1 backbones, followed by expression in mammalian cells and purification by Protein A affinity chromatography. Antibody specificity was evaluated by flow cytometry using HEK293 cells transiently expressing human SLC35D3 (NP_001008783.1), cynomolgus monkey SLC35D3 (XP_005552002.1), or murine SLC35D3 (NP_083805.1), as well as the closely related family members SLC35D1 (NP_055954.1) and SLC35D2 (NP_008932.2). Human IgG1 isotype control (BioLegend, USA) and anti-Myc-tag antibody (clone 9E10; Enzo, USA) were included as staining controls. Epitope mapping was performed using segmental mutants of human SLC35D3 in which amino acid regions were deleted or replaced (see Supplementary Table S1 for details). Cross-reactivity and epitope-mapping assays were conducted at a fixed antibody concentration of 10 μg/mL; binding to endogenously expressing cells was evaluated using SNU-16 WT and isogenic KO derivatives incubated with serial dilutions of antibodies. For all flow-cytometric assays, cells were stained in PBS containing 5% FBS for 30 min at 4°C, washed, and probed with APC-conjugated anti-human IgG secondary antibody (Jackson, USA). Dead cells were stained with FVS450 viability dye (BD Biosciences, USA). Data were acquired on a FACSCantoII (BD Biosciences) and analyzed using FlowJo v10 (BD Biosciences). Mean fluorescence intensity (MFI) values were quantified from FVS450-negative viable populations.
2.3. CAR construct and lentivirus production
The scFv of the humanized anti-SLC35D3 antibody h021-12, derived from the parental #021, was cloned into a second-generation CAR backbone comprising the CD8α hinge-transmembrane region, the 4-1BB costimulatory domain, and the CD3ζ signaling domain. A Flag-TEV tag was fused to the N-terminus of the scFv to facilitate detection and potential proteolytic removal. The CAR cassette was inserted into a pLVSIN-EF1α-pur lentiviral vector (TaKaRa bio, Japan). Lentiviral particles were generated using the Lentiviral High Titer Packaging Mix (TaKaRa bio) in Lenti-X 293T cells (TaKaRa bio) with TransIT-293 Transfection Reagent (Mirus bio, USA) according to the manufacturer's instructions. Viral supernatants were collected on day 2 and day 3 post-transfection, concentrated with Lenti-X Concentrator (TaKaRa bio), and the titers were quantified via Lenti-X p24 Rapid Titer Kit (TaKaRa bio).
2.4. Generation of CAR-T and Control-T cells
Cryopreserved Peripheral Blood Mononuclear Cells (PBMCs) were purchased from CTL (Cellular Technology Limited, USA), with donor consent and ethics compliance being ensured by the supplier. PBMCs (1 × 106 cells/well) and supernatant that contained lentivirus encoding the h021-12 CAR were co-seeded onto 24-well plates pre-coated with anti-CD3 antibody at 2 μg/mL (clone OKT3; BioLegend, USA) and RetroNectin at 10 μg/mL (Takara Bio) in AIM-V medium supplemented with 5% heat-inactivated FBS containing 600 IU/mL recombinant human IL-2 (Nipro, Japan), allowing simultaneous activation and transduction. Two sequential spinoculations (centrifugation at 1000 × g, 30 min, 32°C) were performed 24 h apart using a multiplicity of infection (MOI) of 2 to enhance transduction efficiency. Control-T cells were subjected to the same activation and culture procedures but without viral supernatant. Cells were expanded for 11 days in AIM-V medium supplemented with 10% heat-inactivated FBS and 600 IU/mL recombinant human IL-2, with medium exchange every 2–3 days. CAR expression was verified by flow cytometry (MACSQuant Analyzer 10, Miltenyi, USA) using anti-FLAG antibody (clone L5; BioLegend, USA) and anti-CD3 (clone UCTH1; BioLegend, USA) in the same staining panel, with 7-AAD to exclude non-viable cells. Flow-cytometric data were analyzed using FlowJo v10. The proportions of CAR+ and CD3+ cells within the viable population are summarized in Supplementary Table S2.
2.5. In vitro cytotoxicity assay
GFP-labelled QGP-1 cells (WT/KO/OE) were harvested using TrypLE™ Express Enzyme (Thermo Fisher Scientific) and seeded at 5 × 103 cells per well in flat-bottom 96-well plates. Antigen expression of these target cells was confirmed by flow cytometry using anti-SLC35D3 antibody #021 as described under Antibody generation and profiling. After allowing the cells to adhere, effector cells (Control-T cells or h021-12 CAR-T cells) were added at effector-to-target (E:T) ratios ranging from 0.37:1 to 30:1. Target-only wells and effector-only wells containing the number of effector cells corresponding to the highest E:T ratio were included as controls. Co-cultures were monitored using the Incucyte S3 live-cell imaging system (Sartorius, USA) with images acquired every 3 to 4 h for up to 257 h. Viable tumor cells were quantified as GFP-positive area (μm2/image) using the Incucyte analysis software. Timepoints at 168 h are shown in main figures; extended data (up to 257 h) are in Supplementary Fig. S1. Each condition was assayed in three technical replicate wells within a single independent experiment, using CAR-T cells derived from a single PBMC donor; independent experiments with additional donors were not performed. Data represent mean ± standard deviation (SD) of the three technical replicate wells.
2.6. Cytokine production
Co-culture of target cells (HCT 116 Neg, QGP-1 OE) and effector cells, and confirmation of their antigen expression was conducted as described above. After 24 h of co-culture, cell-free supernatants from the three technical replicate wells of each condition were pooled, and the pooled sample was measured in duplicate using the AlphaLISA Human IFN-γ Detection Kit (PerkinElmer, USA) according to the manufacturer's instructions. Data are presented as the mean of duplicate measurements of the pooled sample, without error bars. Because replicates were pooled prior to measurement, no statistical test was applied and these data are presented as descriptive evidence of antigen-dependent cytokine release.
2.7. In vivo tumor engraftment inhibition study
All animal experiments were approved by the Institutional Animal Care and Use Committee of Daiichi Sankyo Co., Ltd. (approval number: B13039, date: April 6, 2020) and conducted in an AAALAC International-accredited facility. All procedures were performed in compliance with relevant laws, institutional guidelines and ARRIVE guidelines. Female NOD. Cg-PrkdcscidIl2rgtm1Sug/ShiJic (NOG) mice (6 weeks of age; In-Vivo Science, Japan) were purchased and maintained under specific pathogen-free conditions with ad libitum access to food and water.
On day 0, QGP-1 cells (WT/KO/OE as indicated) were harvested using TrypLE™ Express Enzyme, resuspended in a 1:1 mixture of PBS and Matrigel (Corning, USA), and inoculated subcutaneously into the right flank at 1 × 106 cells per mouse. On the same day, T-cell products were administered intravenously via the tail vein in 200 μL PBS. The CAR-T group received 1 × 107 CAR-positive viable cells per mouse, whereas the Control-T group received a total viable-cell dose matched to the CAR-T group. A PBS-only group was included as a reference control. Tumor engraftment and growth were monitored twice weekly using a digital caliper, and tumor volume was calculated as (length × width2)/2. Mice were euthanized with CO2 gas upon reaching humane endpoints or when tumor volume exceeded 2000 mm3.
Primary comparisons were made between the CAR-T and Control-T groups. Group sizes were n = 3 (PBS), n = 5 (Control-T) and n = 5 (h021-12 CAR-T) for both the OE and WT models, and n = 3 per group for the KO model. Data normality was assessed with the Shapiro-Wilk test, and group differences were analyzed using Welch's t-test (unpaired, two-tailed). The PBS group served as a descriptive reference and was not included in statistical comparisons. Body weights remained stable throughout the study and were therefore not subjected to statistical testing. All analyses were performed using GraphPad Prism v9 (Dotmatics, USA), and results are presented as mean ± SE for tumor volume and mean ± SD for body weight.
3. Results
3.1. Generation and characterization of Anti-SLC35D3 antibodies
Chimeric monoclonal antibodies against SLC35D3 were generated by rat DNA immunization followed by hybridoma screening. Two representative clones, #021 and #114, were expressed as human IgG1-constant region chimeric antibodies. Subsequently, flow-cytometric analyses using HEK293 cells transiently expressing human, cynomolgus monkey, or mouse orthologs of SLC35D3, as well as related family members SLC35D1 and SLC35D2, revealed broad cross-species reactivity and high target specificity (Fig. 1A). Epitope mapping with a panel of SLC35D3 mutants carrying deletions or amino acid substitutions confirmed that #021 and #114 bind to non-overlapping epitopes (Fig. 1B). Isogenic SNU-16 cell pairs (WT vs. KO) further validated the specificity of antibodies #021 and #114, with concentration-dependent binding observed only in SLC35D3-expressing cells (Fig. 1C).
Fig. 1.
Profiling of newly generated anti-SLC35D3 antibodies. (A) Cross-species and family reactivity. HEK293 cells transiently transfected with SLC35D3 orthologs (human: hSLC35D3, cynomolgus monkey: cySLC35D3, mouse: mSLC35D3) or two closely related family members, human SLC35D1 and human SLC35D2, were analyzed by flow cytometry. Six histograms are overlaid per panel. Four panels show staining with isotype control, anti-SLC35D3 antibody clone #021, clone #114, or anti-Myc-tag antibody. (B) Epitope mapping by mutant panel. HEK293 cells transiently expressing a panel of SLC35D3 mutants—generated by replacing or deleting short amino-acid segments—were analyzed by flow cytometry. Histograms are shown for staining with isotype control, #021, and #114. (C) Isogenic SNU-16 cells (WT and KO) were incubated with serial dilutions of antibody #021 or #114, with an isotype control. Graphs show MFI versus antibody concentration for each cell type.
3.2. Construction and validation of the h021-12 CAR-T cells
The anti-SLC35D3 antibody #021 was humanized to generate h021-12, which was subsequently incorporated into a second-generation CAR construct comprising an anti-SLC35D3 scFv linked to a CD8α hinge and transmembrane domain, a 4-1BB co-stimulatory motif, and a CD3ζ signaling domain (Fig. 2A). Antibody #114 was not advanced due to amino acid sequence-related developability concerns. Flow-cytometric analysis confirmed robust h021-12 CAR surface expression, with approximately 80% of viable cells being CAR-positive (Supplementary Table S2).
Fig. 2.
Design and functional validation of anti-SLC35D3 CAR-T cells (clone h021-12). (A) Schematic representation of the h021-12 CAR construct, comprising an anti-SLC35D3 scFv fused to a CD8α hinge-transmembrane domain, 4-1BB costimulatory domain, and CD3ζ signaling domain. (B) Flow-cytometric analyses of SLC35D3 surface expression in the three tumor cell lines used for functional assays: the SLC35D3-positive QGP-1 WT cells, KO derivative, and OE counterpart. Red histograms denote isotype controls. The MFI ratio relative to the isotype control is indicated in the upper right corner of each histogram. (C) Cytotoxicity of h021-12 CAR-T cells at 168 h. Viable tumor cells were monitored by Incucyte live-cell imaging. The y-axis indicates GFP-positive cell area (μm2/image). Data are presented as mean ± SD of three technical replicate wells per condition at the indicated effector-to-target (E:T) ratios (x-axis), from a single independent experiment. No statistical test was applied.
3.3. Antigen-dependent cytotoxicity of h021-12 CAR-T cells in vitro
For functional assays, we generated QGP-1 isogenic models with varying SLC35D3 expression confirmed by flow cytometry: KO (no expression), WT (moderate), and OE (high) (Fig. 2B). The cytotoxic potential of h021-12 CAR-T cells was assessed using Incucyte live-cell imaging over 168 h. In KO targets, minimal cytotoxicity was observed, with GFP area comparable between h021-12 CAR-T and Control-T cells. By contrast, when co-cultured with the WT or OE lines, h021-12 CAR-T cells induced pronounced, effector-to-target (E:T) ratio-dependent killing relative to Control-T cells, thereby confirming antigen-dependent cytotoxicity (Fig. 2C). Extended analysis up to 257 h confirmed sustained antigen-dependent activity (Supplementary Fig. S1). Representative GFP fluorescence images of the GFP-labelled target cell lines are shown in Supplementary Fig. S3.
3.4. Cytokine release of CAR-T cells upon antigen encounter
Interferon-γ (IFN-γ) secretion was quantified after 24 h of co-culture at various E:T ratios. h021-12 CAR-T cells released IFN-γ in an E:T ratio-dependent manner when co-cultured with SLC35D3-positive QGP-1 OE cells (reaching ∼2400 pg/mL at 30:1), but not when co-cultured with SLC35D3-negative HCT 116 cells (Fig. 3A and B). Control-T cells produced only background levels regardless of target antigen expression.
Fig. 3.
Cytokine release by h021-12 CAR-T cells. (A) Flow-cytometric analyses of SLC35D3 surface expression on two tumor cell lines: HCT 116, endogenously SLC35D3-negative (Neg) and QGP-1 OE (as described in Fig. 2B). Red histograms indicate isotype controls. (B) h021-12 CAR-T or control T cells were co-cultured with target cells for 24 h as in cytotoxicity assays. Culture supernatants from three technical replicate wells per condition were pooled and analyzed by AlphaLISA (pg/mL, y-axis). Each pooled sample was measured in duplicate, and the mean values are shown as bars without error bars. Negative values were set to zero for display. Blue = control T cells; red = h021-12 CAR-T cells. No statistical test was applied because replicates were pooled before measurement; these data are descriptive.
3.5. In vivo antitumor activity and tolerability in xenograft models
NOG mice were implanted subcutaneously with QGP-1 cells (OE, WT, or KO) and immediately received PBS, Control-T, or h021-12 CAR-T cells intravenously. In the OE model, h021-12 CAR-T cells profoundly suppressed tumor engraftment and growth compared with controls. Tumor volumes in the CAR-T group remained near baseline for approximately 32 days post-implantation, whereas Control-T and PBS groups showed progressive expansion, reaching mean volumes of ∼900 mm3 and ∼1600 mm3, respectively (Fig. 4A). Statistical analyses confirmed a significant difference between CAR-T and Control-T groups at the study endpoint (P = 0.0023). Body weights were stable and comparable among all groups, indicating good tolerability. By contrast, in the KO model, tumor growth proceeded similarly across CAR-T, Control-T, and PBS groups, confirming that antitumor activity was antigen-dependent (Fig. 4B). The WT xenograft model showed only a modest trend toward tumor growth delay in the CAR-T group, which did not reach statistical significance (Supplementary Fig. S2).
Fig. 4.
In vivo tumor-engraftment-inhibition studies: antitumor activity and tolerability of h021-12 CAR-T cells. (A) QGP-1 OE cells were inoculated subcutaneously on day 0, followed immediately by intravenous administration of PBS, Control-T cells, or h021-12 CAR-T cells (engraftment-inhibition design). Tumor volume (mean ± SE) and body weight (mean ± SD) were monitored over time (n = 3 mice for PBS; n = 5 for Control-T; n = 5 for h021-12 CAR-T). Tumor volumes of the h021-12 CAR-T and Control-T groups were compared at the study endpoint using an unpaired two-tailed Welch's t-test (P = 0.0023). The PBS group was included as a descriptive reference and was not included in the statistical comparison. Body weight was not subjected to statistical testing. (B) QGP-1 KO cells were inoculated and treated as in (A). Tumor volume (mean ± SE) and body weight (mean ± SD) were monitored over time (n = 3 per group). Tumor volumes of the h021-12 CAR-T and Control-T groups were compared at the study endpoint using an unpaired two-tailed Welch's t-test (P = 0.2166; not significant). Body weight was not subjected to statistical testing.
4. Discussion
Building on our earlier findings [8], we generated anti-SLC35D3 antibodies and developed h021-12 CAR-T cells, providing initial proof of concept for SLC35D3-targeted CAR-T therapy. Clones #021 and #114 recognized non-overlapping epitopes and cross-reacted with human, cynomolgus monkey, and murine SLC35D3. Murine cross-reactivity was relevant to interpreting overt tolerability in mice, although definitive safety assessment was not possible. Clone #021 was selected for CAR development based on its favorable biophysical properties.
h021-12 CAR-T cells showed antigen-dependent cytotoxicity and IFN-γ release in vitro and suppressed OE, but not KO, tumor growth in vivo. Activity against WT tumors was modest and non-significant, consistent with previous reports that CAR activation depends on surface antigen density [11,12], suggesting that h021-12 CAR-T cells can discriminate targets based on antigen abundance.
Several limitations should be acknowledged. First, the in vivo experiments used a tumor engraftment inhibition model with concurrent CAR-T and tumor cell administration, which does not fully reflect established tumor kinetics or the complexity of the TME which can influence CAR-T persistence and efficacy; delayed-treatment studies in established tumors will be required. Second, h021-12 CAR-T cells showed potent activity against OE tumors but modest effects on WT tumors in vivo, likely because endogenous SLC35D3 expression was below the activation threshold. Consistent with this, IFN-γ secretion at E:T ratio 1:1 was relatively low compared to typical levels (several thousand to tens of thousands of pg/mL) reported for second-generation CAR-T cells [13]. These findings suggest that h021-12 CAR-T cells may require higher antigen density for full activation, highlighting the need for optimization of CAR design or T-cell manufacturing conditions to enhance antigen sensitivity while maintaining tumor selectivity. Third, the present study did not assess CAR-T expansion, persistence, or tumor infiltration in vivo, or the phenotype of the CAR-T product, including CD4/CD8 composition and memory or exhaustion status. Such analyses will be important to understand the limited activity in WT tumors and to guide future optimization. Fourth, the binding affinity of the h021-12 scFv was not determined quantitatively, as SLC35D3 is a multi-pass membrane protein for which a properly folded recombinant antigen suitable for SPR could not be prepared; binding was instead assessed in a cell-based format. Finally, comprehensive histopathological evaluation of normal mouse tissues and examination of the tumors themselves were not performed; therefore, definitive conclusions regarding on-target/off-tumor safety cannot be drawn.
Potential optimization strategies include affinity tuning to adjust the antigen-density threshold [14,15], modification of hinge or spacer length [16], alternative costimulatory or innate-signaling domains [17,18], and armored CAR-T approaches to enhance proliferation and tumor infiltration [19]. However, because the mechanisms underlying the limited efficacy in WT tumors remain undefined, these approaches represent hypothesis-driven directions for future investigation.
CRediT authorship contribution statement
Shunsuke Someya: Writing – original draft, Validation, Methodology, Investigation, Conceptualization. Ming Ming: Validation, Methodology, Investigation. Wataru Tomisato: Supervision, Project administration, Methodology. Keisuke Fukuchi: Writing – review & editing, Supervision, Methodology. Toshiaki Ohtsuka: Writing – review & editing, Project administration, Methodology. Reimi Kawaida: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Investigation, Conceptualization.
Ethics declaration
This study was conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. This study was approved by the Institutional Animal Care and Use Committee of Daiichi Sankyo Co., Ltd. (Approval No. B13039)
Declaration of generative AI and AI-assisted technologies in the manuscript preparation process
During the preparation of this work, the authors used ChatGPT (OpenAI, San Francisco, CA, USA) and DeepL (DeepL SE, Cologne, North Rhine-Westphalia, Germany) to improve the readability and language quality of the manuscript. They were used for English language editing, grammar checking, and literature search assistance. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.
Funding
This study was funded by Daiichi Sankyo Co., Ltd.
Declaration of competing interest
All authors are employees and/or shareholders of Daiichi Sankyo Co., Ltd.
Acknowledgements
We thank the late Toshinori Agatsuma of Daiichi Sankyo Co., Ltd. For his invaluable contributions. The authors also thank Chigusa Yoshimura for generating the antibody, Hayami Sugiyama for the CAR-T production, and Hirokazu Ishikawa for the establishment of the OE/KO cell lines. We acknowledge the contributions of Yusuke Shuchi and Chikako Suzuki with whom we previously collaborated, for their support and insightful input during the project.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbrep.2026.102810.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
figs1.
figs2.
figs3.
Data availability
Data will be made available on request.
References
- 1.June C.H., O'Connor R.S., Kawalekar O.U., Ghassemi S., Milone M.C. CAR T cell immunotherapy for human cancer. Science. 2018;359:1361–1365. doi: 10.1126/science.aar6711. [DOI] [PubMed] [Google Scholar]
- 2.Maciocia N., Wade B., Maciocia P. CAR T-cell therapies for T-cell malignancies: does cellular immunotherapy represent the best chance of cure? Blood Adv. 2025;9:913–923. doi: 10.1182/bloodadvances.2023012263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sterner R.C., Sterner R.M. CAR-T cell therapy: current limitations and potential strategies. Blood Cancer J. 2021;11:1–11. doi: 10.1038/s41408-021-00459-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zugasti I., Espinosa-Aroca Lady, Fidyt K., Mulens-Arias V., Diaz-Beya M., Juan M., Urbano-Ispizua Á., Esteve J., Velasco-Hernandez T., Menéndez P. CAR-T cell therapy for cancer: current challenges and future directions. Signal Transduct. Targeted Ther. 2025;10:1–51. doi: 10.1038/s41392-025-02269-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Du B., Qin J., Lin B., Zhang J., Li D., Liu M. CAR-T therapy in solid tumors. Cancer Cell. 2025;43:665–679. doi: 10.1016/j.ccell.2025.03.019. [DOI] [PubMed] [Google Scholar]
- 6.Li X., Zhou J., Zhang W., You W., Wang J., Zhou L., Liu L., Chen W.-W., Li H. Pan-cancer analysis identifies tumor cell surface targets for CAR-T cell therapies and antibody drug conjugates. Cancers. 2022;14:5674. doi: 10.3390/cancers14225674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yi X., Zhao H., Hu S., Dong L., Dou Y., Li J., Gao Q., Zhang B. Tumor-associated antigen prediction using a single-sample gene expression state inference algorithm. Cell Rep. Methods. 2024;4 doi: 10.1016/j.crmeth.2024.100906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Someya S., Inoue T., Kawaida R., Ohtsuka T., Fukuchi K. Integrated transcriptomic and proteomic validation identifies SLC35D3 as a tumor-selective surface antigen for colorectal and neuroendocrine carcinomas. Biochem. Biophys. Rep. 2026;46 doi: 10.1016/j.bbrep.2026.102587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.van der Stegen S.J.C., Hamieh M., Sadelain M. The pharmacology of second-generation chimeric antigen receptors. Nat. Rev. Drug Discov. 2015;14:499–509. doi: 10.1038/nrd4597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Suzuki H., Nagase S., Saito C., Takatsuka A., Nagata M., Honda K., Kaneda Y., Nishiya Y., Honda T., Ishizaka T., Nakamura K., Nakada T., Abe Y., Agatsuma T. Raludotatug deruxtecan, a CDH6-Targeting antibody–drug conjugate with a DNA topoisomerase I inhibitor DXd, is efficacious in human ovarian and kidney cancer models. Mol. Cancer Therapeut. 2024;23:257–271. doi: 10.1158/1535-7163.MCT-23-0287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ramakrishna S., Highfill S.L., Walsh Z., Nguyen S.M., Lei H., Shern J.F., Qin H., Kraft I.L., Stetler-Stevenson M., Yuan C.M., Hwang J.D., Feng Y., Zhu Z., Dimitrov D., Shah N.N., Fry T.J. Modulation of target antigen density improves CAR T-cell functionality and persistence. Clin. Cancer Res. 2019;25:5329–5341. doi: 10.1158/1078-0432.CCR-18-3784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Walker A.J., Majzner R.G., Zhang L., Wanhainen K., Long A.H., Nguyen S.M., Lopomo P., Vigny M., Fry T.J., Orentas R.J., Mackall C.L. Tumor antigen and receptor densities regulate efficacy of a chimeric antigen receptor targeting anaplastic lymphoma kinase. Mol. Ther. 2017;25:2189–2201. doi: 10.1016/j.ymthe.2017.06.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Castella M., Boronat A., Martín-Ibáñez R., Rodríguez V., Suñé G., Caballero M., Marzal B., Pérez-Amill L., Martín-Antonio B., Castaño J., Bueno C., Balagué O., González-Navarro E.A., Serra-Pages C., Engel P., Vilella R., Benitez-Ribas D., Ortiz-Maldonado V., Cid J., Tabera J., Canals J.M., Lozano M., Baumann T., Vilarrodona A., Trias E., Campo E., Menendez P., Urbano-Ispizua Á., Yagüe J., Pérez-Galán P., Rives S., Delgado J., Juan M. Development of a novel Anti-CD19 chimeric antigen receptor: a paradigm for an affordable CAR T cell production at academic institutions. Mol. Ther., Methods Clin. Dev. 2019;12:134–144. doi: 10.1016/j.omtm.2018.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Vander Mause E.R., Baker J.M., Dietze K.A., Radhakrishnan S.V., Iraguha T., Omili D., Davis P., Chidester S.L., Modzelewska K., Panse J., Marvin J.E., Olson M.L., Steinbach M., Ng D.P., Lim C.S., Atanackovic D., Luetkens T. Systematic single amino acid affinity tuning of CD229 CAR T cells retains efficacy against multiple myeloma and eliminates on-target off-tumor toxicity. Sci. Transl. Med. 2023;15 doi: 10.1126/scitranslmed.add7900. [DOI] [PubMed] [Google Scholar]
- 15.Liu X., Jiang S., Fang C., Yang S., Olalere D., Pequignot E.C., Cogdill A.P., Li N., Ramones M., Granda B., Zhou L., Loew A., Young R.M., June C.H., Zhao Y. Affinity-tuned ErbB2 or EGFR chimeric antigen receptor T cells exhibit an increased therapeutic index against tumors in mice. Cancer Res. 2015;75:3596–3607. doi: 10.1158/0008-5472.CAN-15-0159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Li N., Quan A., Li D., Pan J., Ren H., Hoeltzel G., de Val N., Ashworth D., Ni W., Zhou J., Mackay S., Hewitt S.M., Cachau R., Ho M. The IgG4 Hinge with CD28 transmembrane domain improves VHH-based CAR T cells targeting a membrane-distal epitope of GPC1 in pancreatic cancer. Nat. Commun. 2023;14:1986. doi: 10.1038/s41467-023-37616-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Carpenito C., Milone M.C., Hassan R., Simonet J.C., Lakhal M., Suhoski M.M., Varela-Rohena A., Haines K.M., Heitjan D.F., Albelda S.M., Carroll R.G., Riley J.L., Pastan I., June C.H. Control of large, established tumor xenografts with genetically retargeted human T cells containing CD28 and CD137 domains. Proc. Natl. Acad. Sci. 2009;106:3360–3365. doi: 10.1073/pnas.0813101106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mikolič V., Pantović-Žalig J., Malenšek Š., Sever M., Lainšček D., Jerala R. Toll-like receptor 4 signaling activation domains promote CAR T cell function against solid tumors. Mol. Ther. Oncol. 2024;32 doi: 10.1016/j.omton.2024.200815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Adachi K., Kano Y., Nagai T., Okuyama N., Sakoda Y., Tamada K. IL-7 and CCL19 expression in CAR-T cells improves immune cell infiltration and CAR-T cell survival in the tumor. Nat. Biotechnol. 2018;36:346–351. doi: 10.1038/nbt.4086. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data will be made available on request.