Supplementary Table S1. Homology of SALL4-derived S9V peptide sequence in human proteome

SALL4 induction in cancer cell by 5-azacytidine modulates SALL4-TCR T cells activation. (A) Expression of SALL4 detected by qRT-PCR (top) and Western blotting (bottom) (Sall4 Antibody (G-3): sc-166033 Santa Cruz) for MDA-MB231 treated with 5-azacytidine (5-aza) (2µM) for 72h and 6 days. (B) Degranulation (CD107a) stainings assessed by flow cytometry comparing Mock and SALL4-TCR T cells for their SALL4 reactivity upon coculture with MDA-MB231 treated 72h and 6 Days with 5-Aza. T cells were cocultured for 5h at E:T ratio 1:50 for n=3 different HDs, mean ±SEM. Anti-HLA-A2 (BB7.2) were used for specific reactivity controls. Statistics were calculated using ANOVA test, *p<0.05, ***p<0.001. (C) Relative mRNA expression levels of SALL4 in cancer cell lines pretreated or not with 5-aza (2µM) for 72h. (D) Intracellular TNF-α staining assessed by flow cytometry analysis comparing Mock and SALL4-TCR T cells for their SALL4 reactivity upon coculture with cancer cell lines treated 48h before the coculture with IFN-γ (200UI/mL) and 72 h with 5-aza (2µM) or not. T cells were cocultured for 16h at E:T ratio 1:20. (E) Anti-MHC-I (W6/32) or isotype antibody were used for specific reactivity controls. Cell lines were not treated prior the coculture. HeLa, A2780 and HUH-7 were transduced to express HLA-A2.

SALL4-TCR T cells do not recognize human hematopoietic stem cells CD34+ in vivo. (A) Schematic outlining the experimental procedure of adoptive T cell transfer in hNSG mice. hNSG mice were engrafted IV with 107 Mock (n=9) or SALL4-TCR transduced T cells (n=10) or not (n=8). Seven days after injection, mice were sacrificed and BM were collected. Blood samples were harvested at day 0 before T cells injection and at day 7 following T cell injection. (B) Flow cytometry gating strategy in BM (left) and in the blood samples (right). (left) HSC were determined by the expression of CD34 and CD38 among human CD45 cells. Dendritic cell (DC), in pink, were determined by the medium expression on CD33 and medium expression of CD123, HLA-DR and CD303. Monocytes, in blue, were determined by the high expression of CD33, medium expression on HLA-DR, CD123 and CD303. (right) The percentage of humanization is determined by the ratio mCD45/hCD45. Monocytes were selected from human cells (hCD45+) based on the expression of CD33 and CD14. (C) and (D) Flow cytometry analyses of HSC and myeloid-derived differentiated cells in BM (C) and monocytes in blood (D) among hCD45 cells.

Abstract Aberrant expression of the oncogene SALL4 is associated with stemness, a more aggressive cancer phenotype, and reduced patient survival in various tumor types, making SALL4 a potential target for cancer immunotherapy. We conducted a transcriptional analysis of SALL4 expression in colorectal cancer tissues and demonstrated that SALL4 was overexpressed in primary tumors and paired liver metastases. Then, we identified the SALL4-derived S9V peptide as a naturally processed peptide that induced specific CD8+ T-cell responses from the peripheral blood of patients with gastrointestinal cancer, whereas no responses were observed in the peripheral blood of healthy donors. Thereafter, we isolated an SALL4-specific T-cell receptor (TCR) that recognized this peptide in the most common HLA molecule in the Caucasian population, HLA-A2, and used this to develop TCR-engineered T cells. In vitro analysis showed that SALL4 TCR–redirected primary CD8+ T cells exhibited cytotoxic effects against SALL4-expressing tumor cells and produced effector cytokines. In vivo, SALL4-TCR T cells significantly reduced tumor growth and improved the survival of tumor-bearing mice. Moreover, SALL4-TCR T cells displayed no toxicity against hematopoietic stem cells. Thus, we conclude that T cells engineered to express a SALL4-specific TCR have the potential to be effective as immunotherapy for solid cancers and pave the way for further clinical development.

SALL4-TCR redirected T cells recognize and kill S9V preloaded MDA-MB231 cell line (positive control). Representative graphs of Mock or SALL4-TCR T cells cocultured during 2 days with MDA-MB231 cell line preloaded with S9V peptide. (A) In vitro lysis kinetics obtained by red cytotox (left) and green signal (right) measurements on S9V preloaded MDA-MB231 during 2 days coculture in the presence of Mock (orange) or SALL4-TCR T cells (blue) measured by live-cell imaging IncuCyte. E:T ratio 10:1. Data represent mean ± SD of triplicates, statistics were calculated on the 3000 min time point. ***p<0.0001. (B) Representative pictures where green signals come from GFP, red signals come from red cytotoxic dye measured by live-cell imaging IncuCyte. E:T ratio 10:1. Scale bar, 400nm.

T cell recognition of malignant cells is central to cancer immunotherapy. This process is elicited by interactions between T cell receptors (TCRs) and antigenic peptides displayed on major histocompatibility complex molecules. Sequencing technologies enable characterization of genomic, transcriptomic and epigenetic alterations that can give rise to epitopes in cancer cells, alongside TCR repertoire profiling in T cells. An important challenge is to determine which peptides are recognized by T cells and which TCRs mediate this recognition. This Perspective highlights how technological and computational advances have improved epitope predictions, shed light on TCR-epitope recognition and could help leverage TCR repertoires for therapeutic innovations in cancer immunotherapy.

Abstract T-cell recognition of infected and malignant cells is elicited by the binding of heterodimeric T-Cell Receptors (TCRs) to epitopes and both the TCRα and the TCRβ chains play a key role in these interactions. Machine learning tools trained on databases of TCRs recognizing diverse epitopes are useful for identifying epitope-specific TCRs in large TCR repertoire datasets. However, collecting paired TCRαβ sequences to train such tools is associated with significant sequencing costs. Here we demonstrate that unpaired TCRα + TCRβ sequencing of epitope-specific T cells can be used for training TCR-epitope recognition predictors at a much reduced cost compared to standard single-cell TCR sequencing protocols and with no impact on prediction accuracy. Applying this approach to some unseen epitopes used in the IMMREP community benchmark demonstrates improved accuracy compared to both existing machine learning models and AlphaFold3-based predictions.

SALL-TCR T cells recognize natural SALL4+ HLA-A2+ cancer cell lines.

Adoptive cell therapy (ACT) with tumor-infiltrating lymphocytes (TIL) is a powerful candidate immunotherapy. However, the personalized production process that underlies its potential also introduces an element of stochasticity, where the actual target of the TILs is variable across patients and protocols, and remains largely unknown. In this letter, we describe a proof-of-concept application of T-cell receptor (TCR) sequencing in combination with computational tools for annotating epitope-specificity of the TCR repertoire of TIL products as a potent quality control. We highlight the potential of this approach by demonstrating the tumor antigen-specificity of TCR clusters within the TIL product in silico, and validating these responses in vitro. We also demonstrate screening for off-cancer reactivity against viral antigens, and immune monitoring of identified TCR clones.

S9V is an immunogenic peptide in humanized mouse model. (A) The immunogenicity of SALL4-derived peptides (A9V, T9K, V9L, L10L, K9Y and S9V) has been analyzed in vivo in HLA-DRB1*0101/HLA-A*0201 (A2/DR1) humanized mouse model. These are "Knock Out" mice for murine MHC-I and II molecules. Hockpad immunization starts on day 0 with an injection of 100µg of each peptides and with 50µg of an adjuvant, CpG. Two further injections were performed on day 7 and day 14 with 50µg of each peptides and with 30µg of CpG. The mice were sacrificed on day 16 to harvest the popliteal lymph nodes and SALL4-specific immune responses were analyzed. (B) Representation of ex vivo T cell responses analyzed by murine IFN-γ ELISpot assay for one S9V positive mice. (C) Median intensities and (D) frequencies of SALL4-specific immune responses were assessed by IFN-γ ELISpot assay in popliteal lymph nodes of humanized mouse after hockpad immunization. Responses were considered positive when the IFN-γ spot number was ≥10 and the ratio was 2-fold above the background. Only the positive intensities of specific immune responses were indicated.

Abstract Background: Fibrolamellar carcinoma (FLC) is a rare and often lethal liver cancer that primarily affects children and young adults. There is no approved systemic therapy for FLC. The FLC transcriptome is defined by an in-frame fusion of exon 1 of DNAJB1 with exons 2-10 of PRKACA, resulting in expression of the chimeric oncoprotein DNAJ-PKAc. Direct pharmacologic inhibition of this oncoprotein has been infeasible due to unacceptable on-target toxicity. However, the shared, tumor-specific expression of DNAJ-PKAc presents an opportunity for neoantigen-targeted immunotherapy. Methods: We are conducting an ongoing clinical trial of a peptide vaccine targeting DNAJ-PKAc in combination with immune checkpoint inhibitors. We used single-cell sequencing and functional assays to identify DNAJ-PKAc-specific T cell receptors (TCRs) from peptide-expanded peripheral blood cells of trial participants. We characterized their HLA restriction, avidity, cytotoxicity, cytokine profiles, and differentiation phenotypes using co-culture systems. Results: We identified seven CD4+ TCRs that recognize DNAJ-PKAc in the context of HLA-DRB1*13:01 (n=2) or HLA-DRB3*01:01 (n=5). When introduced into healthy donor T cells and co-cultured with tumor cells expressing the corresponding HLA allele, these TCRs exhibited variable functional avidity, cytokine production, differentiation, and cytotoxic activity against DNAJ-PKAc-pulsed targets. Among them, JHU12-TCR2 (HLA-DRB3*01:01-restricted) mediated the strongest cytokine production and tumor cell killing despite only moderate avidity. Consistent with this, JHU12-TCR2 showed the greatest in vivo expansion in the original patient, who achieved a near-complete response to immunotherapy. In contrast, a fusion-specific TCR identified from a non-responder, JHU8-TCR2, that shares sequence similarities and clusters with JHU12-TCR2, demonstrated limited cytotoxicity in vitro. Interestingly, although all TCRs recognized the same antigen in matched donor T cells, they drove divergent helper T cell differentiation programs. For example, JHU12-TCR2 favored Th1/Th17 polarization with minimal Treg induction, whereas JHU8-TCR2 promoted a Th2-biased phenotype. Conclusions: We identified and functionally characterized multiple CD4+ TCRs specific for the DNAJ-PKAc oncoprotein, highlighting their potential utility as the basis for TCR-based therapy in FLC. A clinical trial of JHU12-TCR2-based TCR therapy for patients with FLC and HLA-DRB3*01:01 is in development. These findings also suggest that variation in vaccine-induced TCR repertoires may contribute to differences in clinical response, and that intrinsic biophysical or signaling properties of individual TCRs can shape the differentiation fate of CD4+ T cells expressing them. Citation Format: Kayla J. Bendinelli, Allison M. Kirk, Marina Baretti, Abigail M. Gottschall, Heng-Chung Kung, Jeric Peter Hernandez, Waqar Arif, Julie Nauroth, Jennifer Durham, Christopher Thoburn, Amanda Huff, Neeha Zaidi, Alexei Hernandez, Hassan Jamaleddine, Timothy N. West, Justin McCallen, George Coukos, Alexandre Harari, Challice L. Bonifant, Elizabeth Jaffee, Won Jin Ho, Grégoire Altan-Bonnet, Paul G. Thomas, Mark Yarchoan. Development of T cell receptor (TCR) based cellular therapy for fibrolamellar HCC uncovers role of TCR in CD4 T cell differentiation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 5620.

Expression of SALL4-TCR in Jurkat TCR KO CD8+ T cells via mRNA electroporation. Representative FACS SALL4-teramer staining on Jurkat TCR KO CD8+ T cells electroporated with mRNA encoding α and β subunits for the SALL4-TCR (right) or an irrelevant-TCR (middle). Mock electroporated cells (no mRNA, left) were used for negative control.

SALL4-derived S9V specific CD8+ T clone recognize and kill MDA-MB231-SALL4 cell line in vitro. Mean number of IFN-γ (left) and GZMB (right) spots counted on ELISpot upon 24h coculture of SALL4 specific CD8+ T clone with MDA-MB231-SALL4 with different E:T ratio 10:1, 5:1, 1:1, 1:5 and 1:10. Data represent mean ± SD of duplicates. Responses were considered positive if the number of SFC was > 10 spots/104 T clone and 2 times the background value. MDA-MB231-CTRL cell line transduced with empty lentiviral vector was used as negative control. Anti MHC-I antibody was used for specific T cell restriction reactivation validation.

Aberrant expression of the oncogene SALL4 is associated with stemness, a more aggressive cancer phenotype, and reduced patient survival in various tumor types, making SALL4 a potential target for cancer immunotherapy. We conducted a transcriptional analysis of SALL4 expression in colorectal cancer tissues and demonstrated that SALL4 was overexpressed in primary tumors and paired liver metastases. Then, we identified the SALL4-derived S9V peptide as a naturally processed peptide that induced specific CD8+ T-cell responses from the peripheral blood of patients with gastrointestinal cancer, whereas no responses were observed in the peripheral blood of healthy donors. Thereafter, we isolated an SALL4-specific T-cell receptor (TCR) that recognized this peptide in the most common HLA molecule in the Caucasian population, HLA-A2, and used this to develop TCR-engineered T cells. In vitro analysis showed that SALL4 TCR-redirected primary CD8+ T cells exhibited cytotoxic effects against SALL4-expressing tumor cells and produced effector cytokines. In vivo, SALL4-TCR T cells significantly reduced tumor growth and improved the survival of tumor-bearing mice. Moreover, SALL4-TCR T cells displayed no toxicity against hematopoietic stem cells. Thus, we conclude that T cells engineered to express a SALL4-specific TCR have the potential to be effective as immunotherapy for solid cancers and pave the way for further clinical development.

Immunophenotyping profile of SALL4-TCR T cells. (A) Memory phenotype: TN/TSCM (CD45RA+, CD62L+), TEMRA (CD62L-, CD45RA+), TCM (CD62L+, CD45RA-) and TEM (CD45RA-, CD62L-) subset distribution in unstimulated and S9V stimulated primary T cells redirected to express SALL4-TCR (n=4). (B) t-SNE representation of the expression of extracellular activation markers and immune checkpoint molecules (CD69, HLA-DR, 4-1BB, PD1, LAG3, and TIM3) of unstimulated and stimulated SALL4-TCR T cells. (C) Expression of CD69, HLA-DR, 4-1BB, PD1, LAG3, and TIM3 on unstimulated and stimulated SALL4-TCR T cells. (D) Co-expression of pairs of checkpoint markers (PD1, TIM3, LAG3, and TIGIT). (E) Expression of chemokine receptors CXCR2, CXCR3, CXCR4, CXCR6, CX3CR1 and CCR4 on SALL4-TCR T cells. (F) Quantification of released analytes including TNF-α, IFNγ, granzyme A, granzyme B, IL2, IL4, and perforin assessed by Legendplex analysis in cell culture supernatants of SALL4-TCR T cells previously stimulated or not during 5h with S9V peptide at different concentration [0-10µg/mL] (n = 4).

SALL4-TCR T cells exhibit anti-tumor activity in vivo.

SALL4-TCR T cells recognize natural SALL4+ cancer cell lines in an HLA-A2 restricted manner.

Peritoneal carcinomatosis is a major therapeutic challenge in gynecological and gastrointestinal malignancies, associated with poor quality of life and short survival. Little is known about the tumor microenvironment (TME) of peritoneal carcinomatosis resistant to chemotherapy. Using multiplexed immunohistochemistry (FoxP3, CD163, CD8, CD68, FAP and Cytokeratin) and sensitive T cell receptor (TCR) repertoire analysis, we investigated spatial and longitudinal changes of the TME of 54 biopsies collected from 18 patients with peritoneal carcinomatosis (six ovarian cancer, nine gastric cancer and three pancreatic cancer) treated in the Nab-PIPAC phase IB trial (repeated cisplatin and nab-paclitaxel by pressurized intraperitoneal aerosol chemotherapy [PIPAC]). Overall, tumors were cold with low T cell infiltration. There was high intra- and inter-patient heterogeneity of TCR repertoire and immune cell distribution, with different tumor patterns and TME compositions depending on the primary cancer. Ovarian and gastric cancers were enriched in CD163+ tumor-associated macrophages while pancreatic cancer was enriched in FAP+ cancer-associated fibroblasts and FoxP3+ regulatory T cells. TCR clonality was higher in ovarian cancer, as compared to oesogastric cancer, likely reflecting expansion of tumor reactive T cells. Focusing on ovarian cancer, hyperexpanded T cells were more abundant in the tumor core as compared to matched stroma. Our spatial and longitudinal study brings new insights into the TME of peritoneal carcinomatosis, with implications for the future design of precision medicine in this disease. Laura Grassi, Raphael Genolet, Valentine Du Bois, Manuela Undurraga, Thibaud Koessler, Jean-Christophe Tille, Antonella Diciola, Mariagrazia Di Marco, Noémie Lang, Nicolas Mach, remi Peanne, Eileen Zhao, Nawel Zouggari, Julie Terzic, Catherine Raimond, Patrick Petignat, Christian Toso, Stefan Monig, Martin Hübner, Frederic Ris, Alexandre Harari, Intidhar Labidi-Galy. Spatial and longitudinal characterization of tumor microenvironment of peritoneal carcinomatosis: Nab-PIPAC phase IB trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6541.

Tumor-infiltrating lymphocyte (TIL) therapy has emerged as a transformative approach in cancer immunotherapy, particularly following the recent US Food and Drug Administration (FDA) approval of lifileucel for advanced melanoma. This review synthesizes current insights into the immune correlates and mechanisms underlying the efficacy of TIL therapy, highlighting the pivotal role of tumor immunogenicity, TIL functional states, and the tumor microenvironment (TME). Recent advances in single-cell profiling and biomarker discovery have enabled more precise patient selection and therapy optimization, while novel expansion protocols and engineered TILs are addressing resistance and broadening applicability to non-melanoma tumors. Collectively, these developments underscore the promise of next-generation TIL therapies to revolutionize treatment paradigms across a wider spectrum of solid cancers.

While cancer immunotherapy has primarily focused on CD8 T cells, CD4 T cells are increasingly recognized for their role in antitumor immunity. The HLA-DRB3*02:02 allele is found in 50% of Caucasians. In this study, we screened HLA-DRB3*02:02 patients with melanoma for tumor-specific CD4 T cells and identified robust New York esophageal squamous cell carcinoma 1 (NY-ESO-1)123-137/HLA-DRB3*02:02 CD4 T cell activity in both peripheral blood and tumor tissue. By analyzing NY-ESO-1123-137/HLA-DRB3*02:02-restricted CD4 T cell clones, we uncovered an unexpectedly high cytotoxicity, strong T helper 1 polarization, and recurrent αβ T cell receptor (TCRαβ) usage across patients and anatomical sites. These responses were also present in other NY-ESO-1-expressing cancers. TCRs from these clones, when transduced into primary CD4 T cells, showed direct antitumor efficacy both in vitro and in vivo. Our findings suggest that these TCRs are promising for adoptive T cell transfer therapy, enabling broader targeting of NY-ESO-1-expressing adult and pediatric cancers in clinical settings.