T cell-mediated immune surveillance is critical for cancer control, yet its role in bone marrow malignancies remains poorly understood. Here, we integrate TCR profiling, HLA immunopeptidomics, and functional screening to characterize tumor-reactive T cells in the bone marrow of patients with multiple myeloma (MM) and acute myeloid leukemia (AML). These cells are transcriptionally defined by a conserved effector program distinct from the exhausted phenotype of tumor-reactive T cells in solid cancers. Immunopeptidomic profiling reveals a partially shared antigen landscape enriched for noncanonical peptides driving convergent TCR responses. We develop TFiT (tumor-reactive features in T cells), a transcriptional classifier that identifies these cells and stratifies immunotherapy, but not chemotherapy, response across independent MM and AML cohorts, supporting its specificity for T cell-mediated tumor control. These findings reveal a latent but activatable anti-tumor T cell compartment in bone marrow malignancies and provide a framework for engaging endogenous immunity in MM and AML.

T cell-based immunotherapies have improved cancer treatment, however their efficacy is frequently limited by tumor-intrinsic resistance mechanisms and by the immune suppressive tumor microenvironment, counteracting tumor specific immune responses through impaired antigen presentation and the accumulation of M2-like tumor-associated macrophages. MicroRNAs have emerged as regulators of tumor cell – immune cell interactions, yet the miRNAs’ capacity to coordinate anti-tumor effects across different cell types is poorly understood. This study aimed to investigate whether miR-155-5p and miR-3535 represent regulators capable of driving coordinated functional reprogramming of both, tumor cells and macrophages. Human tumor cell lines of various cancer entities were transfected with miR-155-5p or miR-3535 and the resulting effects on immune checkpoint molecule expression and tumor cell proliferation were assessed on transcriptional and functional level. Similarly, human PBMC-derived M2 polarized macrophages were transfected with the same microRNAs, followed by analysis of the cytokine secretion patterns and by transcriptomic profiling to evaluate the macrophage polarization states and the immune-regulatory pathways involved. Transcription factor activity and pathway enrichment analyses were performed to identify the regulatory mechanisms affected. Both miR-155-5p and miR-3535 reduced expression of the immune checkpoint molecule CD73 (NT5E) in tumor cells, while miR-155-5p additionally suppressed PD-L1 (CD274) expression. At the same time, both microRNAs reprogrammed M2-like macrophages toward a pro-inflammatory M1-like phenotype, characterized by increased TNFα and CXCL10 secretion and the induction of M1-associated gene expression. Transcriptomic analysis revealed activation of STAT1/IRF-driven inflammatory pathways and a reduced activity of ZNF703, representing a transcriptional hub associated with M2 macrophage infiltration and poor prognosis. Both microRNAs increased TAP1 expression, suggesting enhanced antigen-processing capacity. Furthermore, miR-155-5p and miR-3535 exerted anti-proliferative effects across cell lines of various tumor entities. miR-155-5p and miR-3535 modulate tumor-intrinsic immune checkpoint expression and functional macrophage polarization in a coordinated way, linking tumor cell plasticity to neutralization of the immunosuppressive tumor environment. These findings highlight the potential of coordinated microRNA-mediated regulation via different cell types to overcome immune resistance mechanisms in tumors and support further investigation of microRNA-based strategies in cancer immunotherapy.

T cell receptor (TCR)-engineered T cell therapies hold great promise for the future of precision oncology. Prior to clinical application, candidate TCRs must undergo rigorous validation to ensure their safety and efficacy. These efforts are often hampered by the limited availability of primary, tumor biopsies as source of autologous tumor infiltrating T cells (TILs) and tumor cells, particularly when multiple candidate TCRs await validation. In this study, we established a flow cytometry-based protocol allowing simultaneous functional validation of complex TCR-transgenic T cell populations, thereby significantly reducing the amount of obligatory tumor material required. Effector T cells are color-coded with different fluorochrome conjugated CD45-specific monoclonal antibodies prior to coculture, allowing their discrimination in subsequent flow cytometry analysis without compromising specificity or sensitivity of the assay. Notably, combinatorial CD45 labeling did not interfere with T cell function during the assay over a time course of 24 hours. Here, we validated a five-color code that resolves up to 16 T cell populations through dual-color combinatorial labeling. Higher orders of multiplexing can be achieved by implementing additional colors or triple-color coding.

T cells that recognize tumor-specific mutations are crucial for cancer immunosurveillance and in adoptive transfer of TILs or transgenic-TCR T cell products. However, their challenging identification and isolation limits their use in clinical practice. Therefore, novel approaches to isolate tumor-specific T cells are needed. Here, we report the isolation of neoantigen-specific CD8+ T cells from a vaccination site of a metastatic breast cancer patient who received a personalized vaccine. Based on the somatic mutations, potential MHC binding epitopes were predicted, of which 17 were selected to generate a peptide vaccine. Cutaneous biopsies were processed after the fifth vaccination cycle to obtain infiltrating lymphocytes from the vaccination site (VILs). IFNγ ELISpot revealed reactivity to four peptides used in the vaccine. Reactive T cells from VILs were non-overlapping with those detected in the blood and the tumor-microenvironment. ScTCR Seq analysis revealed the presence of a clonotype in VILs that further expanded after a round of in vitro stimulation and validated to be specific against a private mutation, namely NCOR1L1475R, presented in the context of HLA-B * 07:02, with no reactivity to the wild-type peptide. Our study shows, for the first time, that tumor mutation – specific T cells are generated at high frequencies in the vaccination site and can be isolated with standard methods for TCR screening. The easy and safe accessibility of skin biopsies overcomes the major hurdles of current TCR screening approaches and present exciting opportunities for the development of innovative immunotherapeutic strategies.

Table S3. Primer for single cell T cell receptor sequencing.

Supplementary Figure 1: T cell responses in A2.DR1 and DR4 mice after peptide vaccination with potential shared neoepitopes. A. Predicted binding affinity for selected amino acid alterations for HLA-A*0201 determined by NetMHC 4.0. B, C. Quantification of IFN-γ ELISpot using cells isolated from lymph nodes from vaccinated A2.DR1 mice restimulated with the indicated peptide. D. ELISpot of splenocytes from 27-mer CICR215W-vaccinated mice restimulated for 48 h with the indicated 15-mer (see also Supplementary Table 1). B, C, D Data are represented as mean {plus minus} SEM, statistical significance was determined by unpaired two-tailed Student's t-test. E. Intracellular IFN-γ flow cytometric analyses after in vitro re-stimulation of splenocytes from 27-mer CICR215W-vaccinated DR4 mice gated on single, living, and CD3+ cells.

BACKGROUND:Methods to identify and characterize antigen-reactive T cell receptors (TCRs) represent important tools to understand and exploit T cell responses in patients with cancer and beyond. Current methods are hampered by the rarity of individual T cell clones and limited applicability to monitor both major histocompatibility complex (MHC) class I-restricted and MHC class II-restricted responses, hence insufficiently reflecting the entire antigen-reactive repertoire of a patient. To obtain broad and deep insight into polyclonal, antigen-specific TCR repertoires, we developed the 'epitope-specific expansion culture with subsequent identification of TCRs' (ESPEC-SUIT) assay to identify and track antigen-specific TCRs. METHODS:In vitro stimulation of peripheral blood mononuclear cells with (vaccine-targeted) neoantigens was verified in cytokine secretion assays and read-out by TCRβ repertoire sequencing (TCRseq). Candidate antigen-reactive clonotypes were defined by specific expansion in cultures stimulated with relevant antigen, followed by TCR cloning and validation in co-cultures of TCR transgenic effector cells and peptide-presenting targets. Using TCRseq information, candidate and validated clonotypes were traced and characterized in bulk and single-cell repertoire sequencing data of longitudinally collected blood samples and tumor tissue. RESULTS:In a cohort of 32 patients with cancer, we demonstrate that ESPEC-SUIT supports strong, robust and reproducible expansion of CD4+ and CD8+ T cells in response to various antigens. TCRseq revealed highly polyclonal neoepitope-specific T cell responses, which can be further characterized with respect to cross-reactivity, affinity or human leukocyte antigen (HLA) restriction. In a subcohort of 10 patients, we selected 341 ESPEC-SUIT-derived TCRs for cloning and in vitro functional validation from >2000 candidates and confirmed antigen-reactivity for >75%. We exemplify the usefulness of this TCR discovery method for downstream analysis in neoepitope vaccinated patients with glioma, where we found longitudinal changes in candidate TCR frequencies in blood mirroring antigen-specific ex vivo Enzyme-Linked ImmunoSpot (ELISpot) responses. Furthermore, up to 67% of candidates could be detected in on-treatment brain tumor tissue and exhibited gene expression signatures overlapping with clonotypes of confirmed specificity to the vaccine antigen. CONCLUSION:ESPEC-SUIT provides unprecedented insight into highly polyclonal, antigen-specific T cell responses and enables discovery of large numbers of TCRs for a given antigen. It represents an efficient, cost-effective and scalable framework for the interrogation of tumor-reactive T cell responses in patients with cancer.

Supplementary Figure 2: Molecular features of A2.DR1 MHC-humanized gliomas. A. Verification of the CRISPR-Cas9-mediated functional KO of p53 and Nf1 in the A2.DR1 glioma cell line. B. Copy number variations (CNVs) in the A2.DR1 glioma cell line. C. Chromosomal distribution of CNVs D. Representative immunofluorescence image of A2.DR1 glioma orthotopically injected into the brain of adult A2.DR1 mice, 20 days after injection. E. Relative expression of Cic in wild-type and CicR215W overexpressing A2.DR1 glioma cells as determined by RT-qPCR. F: Immunoblot depicting overexpression of the Cic protein in the CicR215W transfected cell line compared to the parental line (WT).

Table S1A. Peptides; Table S1B. NetMHCIIpan 4.0; Table S1C. NetMHC 4.0

Personalized treatment has become a realistic option for tumor patients, accelerated by significantly reduced sequencing costs of tumor genomes and advances in vaccine formulations. The druggability of cancer neo-antigens caused by individual mutations is centered in this effort. We here use an adeno-associated virus (AAV)-based virus-like particle (VLP) platform to compose a neo-antigen-specific protein vaccine that is effective in a murine prevention and treatment setting. Furthermore, we show that CD4+ T cell responses that are provided by the AAV capsid are crucial for effective murine melanoma treatment. To uncover the optimal composition of a peptide vaccine we de-linked major histocompatibility complex (MHC) class II helper peptides from the capsid and formulated an efficient neo-antigen-specific vaccine, which showed the independence of CD4+ T cell response from tumor sequences. The findings are supported by clinical data of neo-antigen-vaccinated tumor patients. Our results punctuate on the significance of MHC class II epitopes for CD8+ T cell responses and suggest a future use of AAVLPs as neo-epitope vaccines in personalized cancer treatments.

T cell-mediated immune surveillance is critical for cancer control, yet its endogenous effectiveness in hematological malignancies remains limited and poorly understood. Here, we integrate single-cell T cell receptor (TCR) profiling, HLA immunopeptidomics and functional antigen mapping to dissect the specificity landscape of bone marrow lymphocytes (BMLs) in multiple myeloma (MM) and acute myeloid leukemia (AML). We identify a rare subset of tumor-reactive T cells that exhibit a stereotyped transcriptional state distinct from bystander and virus-specific populations. Across both malignancies, immunopeptidomic profiling uncovers a partially conserved antigen repertoire enriched for noncanonical peptides, including products of novel or unannotated open reading frames (nuORFs), pseudogenes, and clonotypic immunoglobulin sequences. Several of these epitopes are recurrently presented and associated with convergent TCR responses across individuals. Based on this immune architecture, we develop a TCR-intrinsic fitness model that infers BML tumor specificity from transcriptional cues and stratifies immunotherapy response across three independent patient cohorts. Together, these findings map the latent potential of endogenous anti-tumor immunity in two biologically distinct diseases and provide a framework for decoding and restoring productive immune surveillance of hematological malignancies. Highlights ### Competing Interest Statement M.J.F reports speaker honoraria from Pfizer, Roche and Kerna Ventures and is a consultant for Moonwalk Biosciences. M.P. and E.W.G. are founders of Tcelltech. S.A.C. is a member of the scientific advisory boards of Kymera, PTM BioLabs, Seer and PrognomIQ. S.F. reports consultancy fees from Illumina. The other authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Deutsche José Carreras Leukämie-Stiftung, https://ror.org/00826gz80, 01ZI/2022 Dr. Rolf M. Schwiete Stiftung, https://ror.org/03v7rwj71, 2025-018 Else Kröner-Fresenius-Stiftung, 2025_EKMS.52 Heidelberg University, https://ror.org/038t36y30, ExU 6.1.12

Table S2. Murine primers for evaluation of Cic expression.

Supplementary Figure 3: T cell effector functions of TCR-transgenic T cells targeting CICR215W. A. Venn diagram of shared TCR-β sequences in T cell lines generated from n = 4 CICR215W vaccinated mice. B. Normalized IFN-γ production of different Î{plus minus}-CICR215W TCR candidates transduced in A2.DR1 T cells and re-stimulated with the indicated peptides that were subsequently vaccinated with CICR215W or MOG. C. Normalized IFN-γ production of expanded CT2 TCR-transduced T cells incubated with 20 µg/ml Î{plus minus}PD-1 and Î{plus minus}CTLA-4 or isotypes and restimulated with CICR215W 27-mer peptide. D. Cell trace far red (CTFR) dilution of CT2 TCR-transduced T cells injected in A2.DR1 recipient mice. E. Representative GFP expression in CD3+ T cells isolated from tumor, contralateral hemisphere (cH), cervical lymph node (cLN) or deep cervical lymph node (dcLN) harvested from A2.DR1 glioma-bearing mice 10 days after adoptive transfer of CT2 Î{plus minus}-CICR215W TCR-transduced T cells. F. Kaplan-Meier curve of A2.DR1 glioma-bearing mice after 2 intraventricular adoptive transfers (AT) of Î{plus minus}CICR215W-TCR T cells and Î{plus minus}Flu-TCR T cells and four doses of ICB (250 µg Î{plus minus}PD-1 and 100 µg Î{plus minus}CTLA-4 or isotype). n = 9 for CIC, n = 8 for Flu, n = 8 for CIC+ICB and n = 8 for Flu+ICB.

T cell receptors (TCRs) play a pivotal role in mediating anti-tumor responses through the recognition of cancer antigens. Although well-characterized in solid tumors, their function in hematological malignancies is less understood due to the high abundance of virus-specific bone marrow T cells (BMTCs) and technical challenges to deorphanize human TCRs at scale. Additionally, the clinical relevance of T cells with anti-tumor reactivity remains unclear for multiple myeloma (MM). Here, we aimed to chart BMTC specificities and transcriptional profiles in MM patients and assessed their significance in mediating clinical anti-tumor responses. We collected BM biopsies and blood samples from 62 MM patients. We utilized longitudinal single-cell RNA and V(D)J sequencing to characterize the phenotype and clonality of over 500,000 BMTCs. Additionally, tumor-reactive T cells were identified by screening 187,015 T cells representing 132,501 unique TCR clonotypes for their reactivity against single autologous MM cells using a newly developed microfluidics-based screening method. We deorphanized putative tumor-reactive TCRs using patient-specific peptide libraries of tumor MHC class I and II immunopeptidomes. By synthesizing, cloning, and re-expressing these TCRs, we examined recognition of MM versus healthy patient cells. We further differentiated TCRs that cross-reacted with viral antigens (CMV, EBV, Influenza, SARS-CoV-2) from those that demonstrated bona fide MM-specificity. We found distinct heterogeneity in patient BMTCs, with specific transcriptional signatures linked to anti-tumor functions as opposed to virus recognition or bystander activity. Deeper profiling of T cells with MM-specificity revealed a transcriptionally unique, ITGB1 (CD29)-expressing T cell subset recognizing cancer-associated antigens (CAAs) or peptides from non-canonical translation products (nuORFs). Additionally, we recurrently identified personalized neoantigens within the immunoglobulin (Ig) hypervariable region of MM clones. T cells targeting such Ig-derived antigens were found in 70% of analyzed patients. Further, a minimal epitope derived from cancer/testis antigen 2 (CTAG2) was recognized by TCRs from multiple MM patients spanning four different HLA supergroups. This data suggests the frequent occurrence of both private neoantigens derived from the hypermutated MM-Ig and public, MHC-promiscuous antigens as targets of endogenous anti-myeloma T cell responses. Surprisingly, the clonal expansion of these tumor-specific TCRs, present in only about 1.2% of assayed BMTCs but detectable in all newly diagnosed MM patients, correlated with enhanced clinical responses. To further investigate the clinical relevance of our findings, we developed and benchmarked a transcriptional signature of anti-tumor reactivity tailored to the unique transcriptional phenotype of lymphocytes within the bone marrow environment. Applying this signature to two independent patient cohorts, we found that early detection of tumor-reactive TCRs correlates with improved responses to current SoC induction/consolidation treatment with Daratumumab-VTd (HR = 0.32, p < 0.0001) as well as BCMAxCD3 bispecific T cell engager administration (HR = 0.36, p = 0.0005) in relapsed/refractory MM. To determine if tumor-reactive TCRs are transplanted and persist long-term, we analyzed stem cell grafts from MM patients undergoing autologous stem cell transplantation (ASCT; NCT03617731) and traced individual TCRs up to two years post-ASCT. We found that tumor-reactive TCRs were enriched in these grafts, preferentially transplanted, and maintained long-term persistence, potentially contributing to the therapeutic value of ASCT. Furthermore, the presence of these TCRs in the bone marrow correlated with long-term clinical responses, indicating their role in sustaining anti-myeloma immunity. Here, we systematically profiled and consistently detected T cells with endogenous anti-tumor reactivity in MM patients. Our data suggest that tumor-reactive TCRs play a critical role in mediating anti-myeloma responses and are influenced by both intrinsic properties and microenvironmental factors. Ongoing studies aim to further elucidate how these TCRs contribute to long-term immune surveillance and therapeutic responses in MM, with potential implications for the development of more effective immunotherapies.

Abstract Innovative immunotherapy approaches such as adoptive transfer of chimeric antigen receptor (CAR) T cells or tumor infiltrating lymphocytes (TILs) have shown great success in the treatment of solid tumors and hematological malignancies. Although treatment of multiple myeloma with CAR T cells can induce deep responses, relapses frequently occur due to antigen escape and limited CAR T cell persistence. TCR-engineered T cells may show prolonged persistence in vivo and could mediate sustained antitumor effects. A further benefit of TCR transgenic T cells is the ability to target intracellular antigens that are inaccessible to CAR T cells, expanding the range of potential targets for immunotherapy. In our project, we propose to identify T cell receptors (TCRs) specifically targeting autologous myeloma cells. Tumor-reactive T cells were identified using the Bruker Cellular Analysis Lightning® platform, allowing simultaneous functional analysis of up to 1500 individual T cell/target cell interactions on a chip. Reactive T cells were identified upon detection of secreted cytokines (IFNγ, TNFα, IL2) and measurement of 4-1BB (CD137) surface expression. Tumor-reactive T cells showing various cytokine secretion patterns and 4-1BB expression profiles were detected in each myeloma patient (on average 11.9 T cells out of 1243 cells tested per assay run). Individual tumor-reactive T cells have been isolated and their TCRs were sequenced. TCR sequences of tumor-reactive T cells were mapped to single-cell RNA sequencing data of T cells from the same patiens to reveal a gene expression signature of myeloma-reactive T cells. TCR genes of reactive T cells were cloned and overexpressed in autologous T cells for functional validation and analysis of tumor derived neoepitope specificity. In summary, we present a pipeline allowing identification of myeloma-recognizing T cells and recovery of bona fide tumor-reactive TCRs eligible for patient-individualized T cell therapy. Citation Format: Tim Robin Wagner, Niklas Kehl, Simon Steiger, Michael Kilian, Bruno Schönfelder, Tamara Boschert, Katharina Lindner, Patrick Schmidt, Karsten Rippe, Hartmut Goldschmidt, Marc-Steffen Raab, Michael Platten, Mirco Friedrich, Stefan B. Eichmüller. Identification of tumor-reactive T cell receptors through functional single cell interaction analyses for personalized T cell therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 14.

Peptide-loaded MHC class I (pMHC-I) multimers have revolutionized our capabilities to monitor disease-associated T cell responses with high sensitivity and specificity. To improve the discovery of T cell receptors (TCR) targeting neoantigens of individual tumor patients with recombinant MHC molecules, we developed a peptide-loadable MHC class I platform termed MediMer. MediMers are based on soluble disulfide-stabilized β2-microglobulin/heavy chain ectodomain single-chain dimers (dsSCD) that can be easily produced in large quantities in eukaryotic cells and tailored to individual patients’ HLA allotypes with only little hands-on time. Upon transient expression in CHO-S cells together with ER-targeted BirA biotin ligase, biotinylated dsSCD are purified from the cell supernatant and are ready to use. We show that CHO-produced dsSCD are free of endogenous peptide ligands. Empty dsSCD from more than 30 different HLA-A,B,C allotypes, that were produced and validated so far, can be loaded with synthetic peptides matching the known binding criteria of the respective allotypes, and stored at low temperature without loss of binding activity. We demonstrate the usability of peptide-loaded dsSCD multimers for the detection of human antigen-specific T cells with comparable sensitivities as multimers generated with peptide-tethered β2m-HLA heavy chain single-chain trimers (SCT) and wild-type peptide-MHC-I complexes prior formed in small-scale refolding reactions. Using allotype-specific, fluorophore-labeled competitor peptides, we present a novel dsSCD-based peptide binding assay capable of interrogating large libraries of in silico predicted neoepitope peptides by flow cytometry in a high-throughput and rapid format. We discovered rare T cell populations with specificity for tumor neoepitopes and epitopes from shared tumor-associated antigens in peripheral blood of a melanoma patient including a so far unreported HLA-C*08:02-restricted NY-ESO-1-specific CD8+ T cell population. Two representative TCR of this T cell population, which could be of potential value for a broader spectrum of patients, were identified by dsSCD-guided single-cell sequencing and were validated by cognate pMHC-I multimer staining and functional responses to autologous peptide-pulsed antigen presenting cells. By deploying the technically accessible dsSCD MHC-I MediMer platform, we hope to significantly improve success rates for the discovery of personalized neoepitope-specific TCR in the future by being able to also cover rare HLA allotypes.

Abstract Determinants of invasion and metastasis in cancer remain of great interest to define. Here, we report the definition of miR-339-3p as a novel tumor suppressive microRNA that blocks melanoma cell invasion without affecting cell survival. miR-339-3p was identified by a comprehensive functional screen of a human miRNA mimetic library in a cell-based assay for invasion by the melanoma cell line A375. miR-339-3p was determined as a strong inhibitor of invasion differentially expressed in melanoma cells and healthy melanocytes. MCL1 was defined as a target for downregulation by miR-339-3p, functioning through direct interaction with the 3′ untranslated region of MCL1 mRNA. Blocking miR-339-3p by an antagomiR was sufficient to increase melanoma cell invasion, an effect that could be phenocopied by RNAi-mediated silencing of MCL1. In vivo studies established that miR-339-3p overexpression was sufficient to decrease lung colonization by A375 melanoma cells in NSG mice, relative to control cells. Overall, our results defined miR-339-3p as a melanoma tumor suppressor, the levels of which contributes to invasive aggressiveness. Cancer Res; 76(12); 3562–71. ©2016 AACR.

Fig. S1. Workflow for a functional screening approach to investigate miRNAs impacting on melanoma cell invasion; Fig. S2. Cell viability assay on 97 most effective miRNA candidates accelerating (A) or inhibiting (B) A375 invasion; Fig. S3. Invasive capacity of melanoma cell lines; Fig. S4. miR-339-3p inhibits invasion in various melanoma cell lines; Fig. S5. MCL1 specific siRNA inhibits melanoma cell invasion; Fig. S6. MCL1 protein expression levels determined in 14 melanoma cell lines; Fig. S7. MCL1 protein expression can be downregulated by miR-339-3p.