Modelling anti-tumor immune responses using patient-derived melanoma organoids

Modelling anti-tumor immune responses using patient-derived melanoma organoids | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Modelling anti-tumor immune responses using patient-derived melanoma organoids Kamila Kaminska, Bengt Phung, Jacob Karlström, Martin Lauss, Katja Harbst, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7486809/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Dec, 2025 Read the published version in Cancer Immunology, Immunotherapy → Version 1 posted 10 You are reading this latest preprint version Abstract Immune checkpoint blockade (ICB) therapy can restore T cell function in tumors, but not all patients benefit, and the mechanisms behind this remain unclear. In this study, we used patient-derived organotypic (PDO) cultures from metastatic melanoma to examine transcriptomic and cellular changes following ex vivo T cell stimulation. Genomic and transcriptomic features were preserved during PDO formation, capturing melanoma heterogeneity. PDOs from ICB-responsive patients showed rapid T cell expansion upon T cell stimulation, unlike those from ICB-resistant tissue. Resistant tissue harbored T cells lacking activation and checkpoint markers, suggesting non-tumor-reactive T cells. A T cell-specific transcriptomic score, activated in responsive PDOs, correlated with improved overall and relapse-free survival in metastatic melanoma patients treated with ICB. These findings demonstrate that ex vivo analysis is a viable tool to investigate mechanisms of ICB response and may help identify predictive biomarkers for patient outcome. Cancer melanoma immune response patient-derived organoid PD1 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Melanoma, a highly aggressive form of skin cancer that poses significant challenges in treatment, especially in its metastatic stage. However, during the last decade, treatment with clinical introduction of immunomodulatory therapies has dramatically changed the outcome of melanoma patients with advanced disease 1 . PD1 blocking antibodies are currently the first-line treatment in advanced melanoma given in the adjuvant or neoadjuvant setting. Despite the central role of PD1 blockade in the clinical management of cancer patients the functional consequences are not fully known. PD1 is expressed on activated T cells, B cells and dendritic cells 2-4 . The ligand of PD1, PDL1, is induced on melanoma cells in response to interferon gamma (IFNG), thereby activating a negative feedback loop that inhibits tumor-specific T cells 5 . Blocking PD1 reactivates these T cells allowing tumor-specific killing to occur. However, a deeper understanding of the molecular consequences of T cell activation within tumors is still needed. Recent studies describing the superior clinical efficacy of neoadjuvant ICB compared to adjuvant ICB underscore the need to understand the early immunological changes occurring within tumors 6 . Voabil et al. 7 used patient-derived tissue fragments (PDTF) to analyze early immunological consequences of PD1 blockade. Their analysis demonstrated that expression of several chemokines such as IFNG and CXCL10, in fragments from patients who responded clinically to ICB, irrespective of cancer type 7 . Alternative model systems to PDTFs include patient-derived organoids 8 9 . These models have the advantage of preserving the tumor microenvironment allowing the evaluation of immunomodulatory effects on both tumor and immune cells. Using patient-derived organoids Ou et al. showed that PD1 blockade activated CD8 + T cells and induced tumor cell death confirming the utility of such models in cancer immunology research 10 . However, the use of such ex vivo models to develop new predictive biomarkers and exploring early transcriptional effects of PD1 blockade have not been fully elucidated. In this study, we generated patient-derived organotypic cultures from fresh metastatic melanoma tissue. These cultures preserved several immune and molecular properties ex vivo . By stimulating T cells from both ICB responders and non-responders, we identified distinct transcriptional signatures that were predictive of patient outcomes following ICB therapy. Collectively, we show that patient-derived organotypic cultures is a valuable model for studying anti-tumor immune responses and discovering new predictive biomarkers. Material and methods Patient samples and organotypic cultures Tumor material used to establish patient derived organotypic cultures was acquired from patients undergoing surgical resection of metastatic melanoma at Skåne University Hospital, Kristianstad General Hospital and Helsingborg General Hospital included in BioMEL 11 (ClinicalTrials.gov ID NCT05446155). All patients undergoing metastatic melanoma surgery (years 2021-2023) from whom tissue was obtained provided written informed consent for the collection of tissue and matched normal blood samples for research as approved by the local ethical board (Lund University Ethical Review Board, Dnr. 2013/101). The study adhered to the declaration of Helsinki. Tissue biopsies were processed fresh usually within 1-2 hours from the surgical procedure. Two types of patient-derived organotypic (PDO) cultures were set up from fresh tumor material Semi-solid PDO cultures modified from Vilgelm et al. 12 and air-liquid interface (ALI) cultures as in Neal et al. 13 with modifications (Supplementary information). T cell stimulation in patient derived organotypic cultures Both types of cultures were treated for 1 week with 10 µg/ml nivolumab (Selleckchem) in the presence of 2 μg/ml anti-CD3 (clone HIT3a, BioLegend) and 2 μg/ml anti-CD28 (clone CD28.2, BioLegend) antibodies. Baseline and treated samples were collected for bulk RNA sequencing. Genomic analyses PDO cultures recovered from the matrix and original tumors were processed for RNA and DNA extraction with AllPrep DNA/RNA Mini Kit (Qiagen). Sequencing libraries were prepared using TruSeq Stranded mRNA Library Prep kit (Illumina), pooled and sequenced in a 2x150 bp paired-end setup (Illumina) on NovaSeq 6000 (Illimuna). Bulk RNA sequencing data were processed using HISAT and Stringtie 14 to obtain fragments per kilobase of transcript per million mapped reads (FPKM) values. Variants in BRAF, NRAS and NF1 were called from hisat bam files using VarDict v.1.8.2 15 . The resulting variants were annotated with Annovar 16 . Copy number alterations were inferred using InferCNV (version 1.18.1). An InfercnvObject was constructed from 71 bulk RNA-seq organoid samples pooled together with 15 bulk RNA-seq samples from benign nevi, which were designated as control samples. MiXCR 17 version 3.0.3 was applied to detect T cell receptor clones. VDJtools version 1.2.1 was run on MiXCR output 18 (Supplementary information). Immunostaining Organoids and a fragment of original tumor tissue from each patient were fixed in 10% neutral buffered formalin solution (Sigma) for 24 hours and processed according to standard dehydration protocol and embedded in paraffin blocks. Sections were taken and stained for SOX10 (BioCare Medical, clone BC34), CD20 (Roche, clone L26) and CD8 (Cell Marque Sigma-Aldrich, clone C8/144B). Immunoslides were scanned with NanoZoomer S60 (Hamamatsu) and processed with NDP.view2.8.24 (Hamamatsu Photonics K.K.) (Supplementary information). Flow cytometry analysis PDO cultures were recovered from culture matrix as described above and further dissociated into single cell suspension. Antibodies used in cocktail in FACS buffer (1% FBS in DPBS): anti-CD45-APC-Vio®770 (clone 5B1; Miltenyi), anti-CD3-PE (clone BW264/56; Miltenyi) and anti-CD19-FITC (clone HIB19; BD). Samples were analyzed on FACSMelody Cell Sorter (BD Biosciences), data were analyzed with FlowJo v10.9.0. Single cell RNA sequencing We generated scRNAseq data using Chromium Next GEM Single Cell 3’ Kit (10x Genomics) according to manufacturer’s recommendations. Libraries were sequenced on NovaSeq6000 (Illumina) as per 10x Genomics User Guide. The h5 files were processed and merged using the R package Seurat 4.0.182, the data were reduced to protein-coding genes, translational (RPS/RPL) and mitochondrial genes (MT-), and genes which a maximum count < = 4 were removed. Cells with less than 500 expressed genes were removed. The data were normalized using SCTransform. Harmony embeddings were calculated from the principal components 19 . The data were visualized using UMAP and clusters were identified using Seurat (Supplementary information). Single cell spatial transcriptomics Single cell spatial transcriptomic data were obtained using the CosMx platform (Nanostring, Seattle, WA) as described previously 20 . Stromal cells were distinguished as endothelial cells and fibroblasts. Tumor cells with B2M log-transformed expression >=3 were considered “MHC-I high”, tumor cells without any expression of MITF, TYR, MLANA and PMEL were considered “Melanin-low”, and else were considered “Melanin-high”. External single cell and bulk RNA sequencing data Single cell RNA sequencing data from Pozniak et al. 21 were downloaded as “Entire_TME.Rds” data file. The data were normalized using SCTransform 22 and initial zero counts were restored. PCA was performed and Harmony was used to adjust for 3‘ vs 5‘ library chemistry 19 . Clustering was performed using the Seurat (Seurat 4.3.0) 23 . UMAP was used to visualize the data and cluster identities were manually assigned using key marker gene expression. External bulk RNA sequencing data have been processed as described previously 24 . Data from Riaz et al. 25 and Gide et al. 26 were used (Supplementary information). Results Patient-derived melanoma organotypic cultures In this study, we aimed to develop model systems suitable for testing immunomodulatory drugs in melanoma. We obtained fresh tumor tissue at surgery from 31 patients diagnosed with metastatic melanoma. The cohort consisted of 17 lymph nodes, 11 cutaneous/subcutaneous metastases and three CNS metastases. We established patient-derived organotypic cultures from fresh tumor material with the aim of preserving a diverse tumor microenvironment. Two types of cultures were established: semi-solid patient-derived organotypic cultures 12 and air-liquid interface patient-derived organotypic cultures 13 ( Fig.1a ). Semi-solid PDO cultures were considered successful upon the formation of three-dimensional spheroid structures within the first week of culture while successful air-liquid PDO cultures exhibited spheroids or dense tumor fragments that maintained their structure and integrity over time ( Fig.1b and Fig.S1 ). The overall success rate was 71%, corresponding to successful establishment of material from 22 out of 31 samples using at least one of the two culture methods ( Fig.1c ) . Air-liquid PDO cultures were successfully established in 11 out of 13 samples (85%), including some samples where semi-solid PDO cultures failed. In conclusion, we have successfully generated patient-derived organotypic cultures from different metastatic sites including, the brain, lymph node and subcutaneous skin tissues. Baseline metastasis and baseline PDOs display genomic similarities Next, we generated whole transcriptomic data from tumor biopsy and matched patient-derived organotypic culture using RNA sequencing. Sufficient RNA was collected from 14 PDOs and matched metastatic deposit with different clinical characteristics ( Table 1 ). Using principal component analysis (PCA) we observed that metastasis and matched PDO were broadly similar in the transcriptional space ( Fig.2a ). To identify cell-type specific differences between the experimental settings, we calculated microenvironment and cell populations-counter (MCP) scores. While most MCP signatures showed similar expression levels, monocytic, fibroblastic and endothelial cell signatures were elevated in baseline tumors compared to PDO ( Fig.2b ). Reassuringly, T- and B cell signatures were preserved during organoid formation. Next, we aimed to statistically assess each PDO was more transcriptionally similar to its corresponding metastasis than to unrelated metastases. We calculated the correlation between each PDO and all analyzed metastatic tumor lesions. Matched PDO and metastases showed significantly higher correlation values compared to unmatched pairs, confirming that transcriptional properties are preserved during organotypic culturing ( P = 1.3x10 -8 , Fig.2c ). Next, we generated genome-wide copy number profiles using the RNA sequencing data. Frequent melanoma-specific copy number changes such as loss of chromosomes 9 and 10 were identified across all samples and those were largely preserved from metastatic tumors to the organotypic cultures. Minor differences were found, such as loss of chromosome 11 in the PDO in Patient 9. In the same case, we observed loss of chromosome 15 that was not detected in the PDO. Nevertheless, the transcriptional data were supported by matched PDO and metastases displaying an overall similar DNA copy number profiles ( Fig.2d, Fig. S2 ). This was further supported by somatic mutations in BRAF , NRAS and NF1 genes that were preserved during the formation of organoids ( Fig.2e ). Moreover, a set of five semi-solid PDOs clustered together in the PCA plot ( Fig.2a ). Detailed analysis of DNA copy number data and hot-spot mutations revealed devoid of genetic alterations. Histological analysis using SOX10 staining further confirmed the absence of melanoma cells ( Fig.S2-3 ). In all, these data suggest that these PDOs lack melanoma cells and thus do not represent their matched metastatic lesions. In conclusion, the generation of PDOs preserves important genomic properties of the original metastases. Cellular and histological characteristics of PDOs Next, we evaluated cellular and histological characteristics of baseline PDO and their matched metastasis. First, we used flow cytometry analysis to determine the fraction of CD3 + /CD45 + T cells in the PDOs. The proportion of CD3 + /CD45 + T cells varied widely, ranging from 0.5% to 92%, reflecting the heterogeneity across cultures ( Fig.3a ). In addition, different proportions of CD45 + /CD3 - immune cell populations were detected, suggesting the presence of immune cell types other than T cells. Secondly, we performed immunostaining of PDOs to determine immune cell localization and found a substantial heterogeneity in the presence and infiltration of CD8 + T cells within the organoids ( Fig.3b, Fig. S4 ). Notably, CD8 + T infiltration of the organoid was limited in the majority of PDOs. Visible CD20 + B cells in conjunction with an organoid were only observed in two cases ( Fig.3b ). One case (Patient 7), despite massive infiltration of CD8 + T and CD20 + B cells within the organoids, SOX10 + melanoma cells were still present ( Fig.3c ). The matched metastatic lesion was subsequently analyzed by immunostaining for immune cells presence. Consistent with organoid findings, the metastatic lesion of organoid from Patient 7 displayed extensive CD8 + T cell infiltration with prominent CD20 + B cell clusters resembling tertiary lymphoid structures (TLS) 24 ( Fig.3c ). In contrast, Patient 6 exhibited limited infiltration of CD8 + T cells that was similarly reflected in the corresponding organoid ( Fig.S5 ). Clearly, not all T cells are infiltrating the actual organoid but are part of the organotypic culture as demonstrated in the flow cytometry data ( Fig.3a ). Melanoma cell states are known to shift in response to microenvironmental cues 27 . To investigate this, we performed single cell RNA sequencing (scRNAseq) on the metastatic lesion from Patient 6 and, in parallel, on tumor organoids isolated from the corresponding culture. Melanoma cells were identified based on SOX10 expression and clustering with UMAP was used to define distinct melanoma cell states. Overall, 10.258 melanoma cells were used in the analysis. Previously, Pozniak et al. described melanoma cell states by scRNAseq 21 . To recapitulate these states, we used the geneset described in Pozniak et al. This approach identified five UMAP clusters with cluster 2 enriched for melanoma cells from the metastatic lesion and cluster 1 enriched for melanoma cells from the PDO ( Fig.3d-e ). However, none of the clusters were exclusive to either metastatic lesion or PDO ( Fig.3e ). Moreover, all clusters expressed both melanocytic and antigen-presenting molecules indicating only subtle changes between them and suggesting that overall melanoma cells from the PDO closely resemble those in the metastatic lesion ( Fig.3f ). In conclusion, the presence of immune cells in baseline metastatic lesion is reflected in PDOs, suggesting an intact anti-tumor immune response. Moreover, melanoma cell states are preserved in the PDOs, reflecting those of the original tumors. Transcriptional effect of T cell stimulation is based on clinical response to ICB To test the relevance of PDO for evaluating anti-cancer immune responses, we used four PDOs derived from patients who experienced clinical benefit from PD1 blockade. As controls, we used four PDOs derived from patients who relapsed following either single-agent PD1 blockade or combined PD1/CTLA4 therapy. To determine the early effects of T cell stimulation, we treated PDOs with CD3 and CD28 agonists in combination with a PD1 blocker for 1 week. PDOs were processed for RNA sequencing pre-treatment and one week of treatment ( Fig.4a ). We then applied MCP-counter to infer presence and abundance of cell types within the cultures. Broadly, baseline PDOs from patients having clinical benefit to ICB showed higher expression scores for antigen presentation, while those from ICB resistant metastases exhibited increased fibroblastic signatures. Although PDOs from patients with clinical benefit from ICB showed higher T cell-related gene expression scores, extensive heterogeneity was observed ( Fig.4b ). Next, we determined T cell receptor clonality (TCR) and found a higher frequency of T cell clones in baseline PDOs from ICB responders. In three of these patients T cell stimulation led to clonal expansion, whereas no such expansion of was observed in cultures from non-responders ( Fig.4c ). Moreover, PDOs from responders showed increased T cell diversity following T cell activation ( Fig.S6 ). Histology analyses of the matched metastatic lesions from patients 4, 7 and 12 identified distinct regions of immune clusters resembling tertiary lymphoid structures ( Fig.4d ) suggesting that such melanomas with benefit of ICB are susceptible to T cell reinvigoration. Patients 4 and 12 both received adjuvant PD1 blockade and were recurrence-free after >1 year. Patient 7 in the cohort received adjuvant anti-PD1/LAG3 therapy, which was halted due to toxicity. A recurrence was detected after almost a year after surgery. The patient subsequently received anti-PD1/CTLA4 therapy, which was discontinued after two cycles. At the most recent follow-up (< 6months from treatment pause), no disease progression had been observed. In contrast, histological analysis of metastatic lesion from patient 13 displayed prominent CD8 + T cell presence however spatially localized in clusters ( Fig.4d ). The other three ICB resistant cases all had very few infiltrating CD8 + T cells ( Fig.S7 ). Next, we wanted to understand why there was a T cell expansion in patients with clinical benefit of ICB while this was not observed in samples from patients resistant to ICB. Thus, we leveraged single cell spatial trancriptomics of two metastatic lesions from Patients 7 and 13. In Patient 7 we observed a wide range of different immune cells including activated CD8 + T cells, cells of the monocytic lineage, B cells and other T cells ( Fig.4e ). We also found a significant proportion of melanoma cells expressing MHC class I suggesting that they are actively presenting antigens. Such cells were frequently located in proximity to immune cells while melanoma cells spatially located distant from immune cells had a decreased expression of MHC class I ( Fig.4e ). In contrast, we generated the same type of data from Patient 13 that had relapsed on adjuvant anti-PD1 therapy but still had a substantial number of T cells which did not show clonal expansion ex vivo ( Fig.4c ). We found that this melanoma lacked tumor cells expressing MHC class I molecules and phenotypically such melanoma cells also had decreased melanocytic transcriptional properties. We also observed a massive infiltration of macrophages but also CD8 + T cells ( Fig.4e ). When analysing the transcriptional phenotypes of CD8 + T cells in these two melanomas we found that CD8 + T cells in Patient 7, who had benefit of ICB therapy, had expression of checkpoint molecules such as PD1, LAG3, TIM3 and TIGIT, while CD8 + T cells from Patient 13 completely lacked expression of such checkpoint/activation molecules (P = 0.0007, Fig.4f ). Collectively, these findings indicate that T cells from patients who benefit of ICB are capable of expanding ex vivo and display transcriptional profiles consistent with activation and checkpoint engagement. T cell stimulation in PDOs from therapy naïve melanoma tissue confers a distinct transcriptional profile predictive of ICB clinical response We next determined the gene programs upregulated by T cell stimulation ex vivo in PDOs from patients with clinical benefit from ICB. Intriguingly, we observed strong induction of genes related to T cell activation including, CCL1, OX40, ICOS, CTLA4 and CXCL13 ( Fig.5a ). Consistent with the T cell clonality results, Patients 4 and 7 displayed the strongest effect on gene expression changes. However, similar transcriptional responses were also observed in the other two PDOs from patients who had benefited from ICB. In these ICB responder cases, gene ontology analysis revealed enrichment of general-immune system processes such as lymphocyte and T cell activation were enriched. Intriguingly, these genes remained unaltered in T cell-stimulated PDOs derived from patients who previously had relapsed on ICB ( Fig.5a ). Next, we investigated genes upregulated in PDOs from patients who had relapsed on ICB including Patient 13 whose baseline culture contained a substantial number of T cells ( Fig.4c ). The genes displaying the highest fold change between baseline and stimulated PDOs from ICB-resistant tissue were enriched for innate immune response genes. Notably, this transcriptional increase was predominantly found in a single patient, Patient 13. This patient relapsed 9 months after adjuvant anti-PD1 therapy and underwent surgical resection of a subcutaneous lesion prior to initiating dual PD1/CTLA4 blockade. At eight-month follow-up, the patient remained recurrence-free. Given the consistent upregulation of a transcriptional program in PDOs treated with T cell stimulatory molecules derived from patients with clinical benefit from ICB, we wanted to explore this in detail. Analysis of single cell RNA sequencing data from melanoma metastases 21 revealed that 14 of the upregulated genes were exclusively expressed in T cells ( Fig.5b ). To determine the predictive value of these 14 T cell specific genes, we created a composite score and applied this on bulk RNA sequencing data from two independent datasets 25 26 . In the Riaz et al. 25 RNA sequencing data on pre- and on-treatment biopsies from 42 metastatic melanoma patients were available. Confirming our ex vivo results, we found that the 14-gene T cell signature was increased in on-treatment biopsies from patients with stable disease, partial response or complete response ( Fig.5c ). Such increase in gene expression signature was not observed in patients with progressive disease on ICB ( Fig.5c ). Finally, we used RNA sequencing data from 69 melanoma patients receiving anti-PD1 or anti-PD1/CTLA4 therapy 26 and applied the T cell signature on these samples. Stratifying patients into tertiles based on the T cell gene expression score displayed significant difference in progression-free and overall survival ( Fig.5d , P = 0.01) in patients with highest T cell gene expression score. In conclusion, transcriptomic signature derived from ex vivo melanoma models can be utilized to predict ICB clinical response. Discussion In this study, we generated and validated PDO systems from metastatic melanoma. These models serve as physiologically relevant models to evaluate immunomodulatory therapies. Utilizing fresh tumor tissues from 31 patients, we tested two culture approaches designed to preserve the cellular heterogeneity and spatial organization of the tumor microenvironment (TME). The overall success rate was 71%, with the air-liquid PDOs showing a higher establishment rate (85%). Notably, the latter succeeded even in cases where semi-solid PDOs failed, indicating the robustness of this model for capturing TME complexity ex vivo . These results corroborate previous efforts to develop organotypic culture systems for human cancers 13 . In our study, transcriptomic profiles of the PDOs closely matched their parental metastatic lesions. Although microenvironment-associated gene signatures such as monocytic, fibroblastic, and endothelial components were decreased in PDOs, critical immune cell lineages, including T and B cells, were preserved. These findings are consistent with the notion that organoid cultures can partially retain the TME, especially lymphoid components, which play pivotal roles in mediating response to immunotherapy 28 . Furthermore, matched tumor-PDO comparisons revealed significantly higher transcriptional correlation than unmatched pairs, confirming that PDO largely conserves tumor-intrinsic transcriptional programs. Copy number variation analysis from RNA-seq data also revealed conserved melanoma-specific alterations, including frequent losses of chromosomes 9 and 10, which are well-documented in melanoma pathogenesis 29 . Single-cell RNA sequencing (scRNA-seq) of melanoma cells from matched tumor and PDO identified overlapping transcriptional states, characterized by shared expression of melanocytic markers and antigen-presentation genes. This suggests that key tumor-intrinsic programs are preserved ex vivo . Further confirming that molecular states are preserved during PDO formation immunostaining revealed variable infiltration of CD8 + T cells in the organoids. In one case (Patient 7), substantial infiltration was observed and reflected the matched metastatic lesion. In contrast, Patient 6 exhibited minimal immune infiltration in both the PDO and tumor. These findings support the reliability of PDOs in reflecting the immune landscape of the original tissue, consistent with previous reports in melanoma organoids 10 . To investigate the utility of PDOs in predicting response to ICB therapies, we stimulated organotypic cultures from patients with known clinical benefit from PD1-based therapy, as well as from patients who had relapsed after ICB, using T cell agonists and PD1 blockade. Post-treatment transcriptional data revealed significant upregulation of T cell activation signatures (e.g., CCL1, OX40, ICOS, CTLA4, CXCL13) in PDOs from ICB responders. In contrast, these changes were absent in PDOs derived from ICB-resistant cases. The chemokine CXCL13, for instance, is expressed by activated T cells and other stromal cells and plays a central role to support TLS formation. Previous work has shown its predictive value for ICB responses in melanoma 30 . Clonotype analysis revealed T cell expansion and increased TCR diversity in responder PDOs, but not in resistant cases, suggesting that the ex vivo models recapitulate in vivo immune responsiveness. Indeed, the PDOs showing the strongest T cell expansion following PD1 blockade were derived from matched metastatic lesions harboring tertiary lymphoid structures (TLS). TLS are linked to improved outcomes in melanoma and other cancers due to their role in sustaining localized anti-tumor immune responses 24 31 . To explore the mechanism underlying these observations, we performed single cell spatial transcriptomics on matched tumors samples. In ICB responders, melanoma cells expressing MHC class I were found in proximity to activated CD8 + T cells, consistent with effective antigen presentation and T cell priming. Conversely, in ICB-resistant cases, such as Patient 13, melanoma cells lacked MHC class I expression and melanocytic markers and CD8 + T cells failed to express activation or checkpoint markers (e.g., PD-1, LAG3, TIM3), suggesting a dysfunctional immune microenvironment. These observations are supported by earlier studies in that have linked MHC class I loss and antigen presentation defects with ICB resistance 32-34 . Using the generated bulk RNA sequencing data from stimulated PDOs of ICB responders we derived a 14-gene T cell activation signature. This signature was validated in independent clinical datasets 25 26 . These findings suggest that PDOs, when used in conjunction with T cell stimulation assays, can serve as a functional readout of a patient’s capacity to mount an immune response and may aid in stratifying patients for immunotherapy. In conclusion, our findings establish that patient-derived organotypic cultures from metastatic melanoma retain key genomic, cellular, and immunological features of the original tumor. The ability to simulate T cell activation and monitor clonal expansion and transcriptional changes ex vivo, provides a promising avenue for biomarker discovery and personalized immunotherapy testing. Declarations Acknowledgements We thank research nurses Marie Sjögren, Kerstin Reistad and Carina Eriksson for assisting in tissue handling. We thank Jari Häkkinen for assistance on aligning the RNA sequencing data. We thank Joanna Pozniak at KU Leuven for access to the single cell RNA sequencing data from patients receiving immune checkpoint blockade. We also thank the clinicians at Helsingborg Hospital, Kristianstad Hospital, and Skåne University Hospital for including patients in the BioMEL study. The authors would like to acknowledge Clinical Genomics Lund, SciLifeLab and Center for Translational Genomics (CTG), Lund University, for providing expertise and service with sequencing and analysis. The computations and data handling were enabled by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS) at Uppsala Multidisciplinary Center for Advanced Computational Science (UPPMAX) partially funded by the Swedish Research Council through grant agreement no. 2018-05973. We also thank Nanostring for CosMx experimental analyses. Data availability Requests for information and resources should be directed to and will be fulfilled by the lead contact, Göran Jönsson ( [email protected] ). Raw sequencing data are regarded as personal information, and by Swedish law, they cannot be made publicly accessible. However, information and mechanisms for data access can be obtained by contacting the corresponding author. Funding This work was supported by Swedish Research Council (Vetenskapsrådet Dnr 2018-02786, Dnr 2022-00871, GJ), Swedish Cancer Society (19 0458 Pj, 22 2105 Pj, GJ), Berta Kamprad Foundation (GJ) and the governmental funding for healthcare research (ALF, GJ and KN), Knut and Alice Wallenberg Foundation (KAW 2022.0066, GJ) and Göran Gustafsson Foundation (GJ). KN was supported by the governmental funding for healthcare research (ALF), the S.R. Gorthon foundation, Hudfonden/Welander-Finsen foundation, Research grants of the Southern health care region, the Gyllenstiernska Krapperup foundation and by the Inga and John Hain foundation for medical research. We gratefully acknowledge the support by Mrs. Berta Kamprad's Cancer Foundation to the L2CancerBridge program at CREATE Health Cancer Center. Author contributions K.K. conducted single cell sequencing, immunostaining, data analysis, design of study and writing the manuscript. B.P. conducted all immunostaining and analysis of such data. K.H. conducted RNA-, TCR and single cell RNA sequencing. K.N., T.S., A.C., H.E and K.I. collected clinical information on patients and set up protocols for collection of viable tumor tissue from patients with melanoma. K.P conducted flow cytometry analysis. M.L and J.K conducted data analysis on bulk and single cell RNA sequencing. G.J. supervised and designed study and wrote the manuscript. All authors have read and approved the manuscript. Declaration of interest The authors declare no competing interests. References Wolchok JD, Chiarion-Sileni V, Rutkowski P, et al. Final, 10-Year Outcomes with Nivolumab plus Ipilimumab in Advanced Melanoma. N Engl J Med 2025;392(1):11-22. doi: 10.1056/NEJMoa2407417 [published Online First: 20240915] van der Leun AM, Thommen DS, Schumacher TN. CD8(+) T cell states in human cancer: insights from single-cell analysis. Nat Rev Cancer 2020;20(4):218-32. doi: 10.1038/s41568-019-0235-4 [published Online First: 20200205] Xiao X, Lao XM, Chen MM, et al. PD-1hi Identifies a Novel Regulatory B-cell Population in Human Hepatoma That Promotes Disease Progression. Cancer Discov 2016;6(5):546-59. doi: 10.1158/2159-8290.CD-15-1408 [published Online First: 20160229] Krempski J, Karyampudi L, Behrens MD, et al. Tumor-infiltrating programmed death receptor-1+ dendritic cells mediate immune suppression in ovarian cancer. J Immunol 2011;186(12):6905-13. doi: 10.4049/jimmunol.1100274 [published Online First: 20110506] Garcia-Diaz A, Shin DS, Moreno BH, et al. Interferon Receptor Signaling Pathways Regulating PD-L1 and PD-L2 Expression. Cell Rep 2017;19(6):1189-201. doi: 10.1016/j.celrep.2017.04.031 Blank CU, Lucas MW, Scolyer RA, et al. Neoadjuvant Nivolumab and Ipilimumab in Resectable Stage III Melanoma. N Engl J Med 2024;391(18):1696-708. doi: 10.1056/NEJMoa2402604 [published Online First: 20240602] Voabil P, de Bruijn M, Roelofsen LM, et al. An ex vivo tumor fragment platform to dissect response to PD-1 blockade in cancer. Nat Med 2021;27(7):1250-61. doi: 10.1038/s41591-021-01398-3 [published Online First: 2021/07/10] Jenkins RW, Aref AR, Lizotte PH, et al. Ex Vivo Profiling of PD-1 Blockade Using Organotypic Tumor Spheroids. Cancer Discov 2018;8(2):196-215. doi: 10.1158/2159-8290.CD-17-0833 [published Online First: 2017/11/05] Sun L, Kang X, Ju H, et al. A human mucosal melanoma organoid platform for modeling tumor heterogeneity and exploring immunotherapy combination options. Sci Adv 2023;9(43):eadg6686. doi: 10.1126/sciadv.adg6686 [published Online First: 20231027] Ou L, Liu S, Wang H, et al. Patient-derived melanoma organoid models facilitate the assessment of immunotherapies. EBioMedicine 2023;92:104614. doi: 10.1016/j.ebiom.2023.104614 [published Online First: 20230523] Helkkula T, Christensen G, Ingvar C, et al. BioMEL: a translational research biobank of melanocytic lesions and melanoma. BMJ Open 2024;14(2):e069694. doi: 10.1136/bmjopen-2022-069694 [published Online First: 20240202] Vilgelm AE, Bergdorf K, Wolf M, et al. Fine-Needle Aspiration-Based Patient-Derived Cancer Organoids. iScience 2020;23(8):101408. doi: 10.1016/j.isci.2020.101408 [published Online First: 20200724] Neal JT, Li X, Zhu J, et al. Organoid Modeling of the Tumor Immune Microenvironment. Cell 2018;175(7):1972-88 e16. doi: 10.1016/j.cell.2018.11.021 [published Online First: 2018/12/15] Pertea M, Kim D, Pertea GM, et al. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat Protoc 2016;11(9):1650-67. doi: 10.1038/nprot.2016.095 [published Online First: 2016/08/26] Lai Z, Markovets A, Ahdesmaki M, et al. VarDict: a novel and versatile variant caller for next-generation sequencing in cancer research. Nucleic Acids Res 2016;44(11):e108. doi: 10.1093/nar/gkw227 [published Online First: 20160407] Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 2010;38(16):e164. doi: 10.1093/nar/gkq603 [published Online First: 2010/07/06] Bolotin DA, Poslavsky S, Mitrophanov I, et al. MiXCR: software for comprehensive adaptive immunity profiling. Nat Methods 2015;12(5):380-1. doi: 10.1038/nmeth.3364 Shugay M, Bagaev DV, Turchaninova MA, et al. VDJtools: Unifying Post-analysis of T Cell Receptor Repertoires. PLoS Comput Biol 2015;11(11):e1004503. doi: 10.1371/journal.pcbi.1004503 [published Online First: 20151125] Korsunsky I, Millard N, Fan J, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 2019;16(12):1289-96. doi: 10.1038/s41592-019-0619-0 [published Online First: 20191118] Elewaut A, Estivill G, Bayerl F, et al. Cancer cells impair monocyte-mediated T cell stimulation to evade immunity. Nature 2025;637(8046):716-25. doi: 10.1038/s41586-024-08257-4 [published Online First: 20241127] Pozniak J, Pedri D, Landeloos E, et al. A TCF4-dependent gene regulatory network confers resistance to immunotherapy in melanoma. Cell 2024;187(1):166-83 e25. doi: 10.1016/j.cell.2023.11.037 Hafemeister C, Satija R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression. Genome Biol 2019;20(1):296. doi: 10.1186/s13059-019-1874-1 [published Online First: 20191223] Butler A, Hoffman P, Smibert P, et al. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol 2018;36(5):411-20. doi: 10.1038/nbt.4096 [published Online First: 20180402] Cabrita R, Lauss M, Sanna A, et al. Tertiary lymphoid structures improve immunotherapy and survival in melanoma. Nature 2020;577(7791):561-65. doi: 10.1038/s41586-019-1914-8 [published Online First: 2020/01/17] Riaz N, Havel JJ, Makarov V, et al. Tumor and Microenvironment Evolution during Immunotherapy with Nivolumab. Cell 2017;171(4):934-49 e15. doi: 10.1016/j.cell.2017.09.028 Gide TN, Quek C, Menzies AM, et al. Distinct Immune Cell Populations Define Response to Anti-PD-1 Monotherapy and Anti-PD-1/Anti-CTLA-4 Combined Therapy. Cancer Cell 2019;35(2):238-55 e6. doi: 10.1016/j.ccell.2019.01.003 [published Online First: 2019/02/13] Karras P, Bordeu I, Pozniak J, et al. A cellular hierarchy in melanoma uncouples growth and metastasis. Nature 2022;610(7930):190-98. doi: 10.1038/s41586-022-05242-7 [published Online First: 20220921] Dijkstra KK, Cattaneo CM, Weeber F, et al. Generation of Tumor-Reactive T Cells by Co-culture of Peripheral Blood Lymphocytes and Tumor Organoids. Cell 2018;174(6):1586-98 e12. doi: 10.1016/j.cell.2018.07.009 [published Online First: 20180809] TCGA. Genomic Classification of Cutaneous Melanoma. Cell 2015;161(7):1681-96. doi: 10.1016/j.cell.2015.05.044 [published Online First: 2015/06/20] Becht E, Giraldo NA, Lacroix L, et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol 2016;17(1):218. doi: 10.1186/s13059-016-1070-5 Helmink BA, Reddy SM, Gao J, et al. B cells and tertiary lymphoid structures promote immunotherapy response. Nature 2020;577(7791):549-55. doi: 10.1038/s41586-019-1922-8 [published Online First: 2020/01/17] Zaretsky JM, Garcia-Diaz A, Shin DS, et al. Mutations Associated with Acquired Resistance to PD-1 Blockade in Melanoma. N Engl J Med 2016;375(9):819-29. doi: 10.1056/NEJMoa1604958 [published Online First: 2016/07/20] Sade-Feldman M, Jiao YJ, Chen JH, et al. Resistance to checkpoint blockade therapy through inactivation of antigen presentation. Nat Commun 2017;8(1):1136. doi: 10.1038/s41467-017-01062-w Lauss M, Phung B, Borch TH, et al. Molecular patterns of resistance to immune checkpoint blockade in melanoma. Nat Commun 2024;15(1):3075. doi: 10.1038/s41467-024-47425-y [published Online First: 20240409] Table 1 Table 1. Clinical features of samples where organotypic cultures were included in the transcriptomic analyses. ICB – immune checkpoint blockade Clinical feature Fraction Gender Male Female N=11, 79% N=3, 21% BRAF mutation (clinical testing) V600E Wt NA N=4, 29% N=8, 57% N=2, 14% Primary tumor site Lower extremity Upper extremity Trunk HN Unknown primary NA N=2, 14% N=1, 7% N=1, 7% N=4, 29% N=5, 36% N=1, 7% Primary tumor type SSM NM Mucosal Unknown primary N=7, 50% N=1, 7% N=1, 7% N=5, 36% Primary tumor Breslow 5.8 mm, range 2.2-14 mm Metastasis LN SC CNS N=10, 71% N=3, 21% N=1, 7% Stage at metastasis IIIB IIIC IIID M1a M1d N=3, 21% N=6, 43% N=2, 14% N=2, 14% N=1, 7% Sampling with regards to treatment Pre-ICB treatment Relapse after ICB On-ICB treatment Therapy naïve NA N=6, 43% N=5, 36% N=1, 7% N=1, 7% N=1, 7% Additional Declarations No competing interests reported. Supplementary Files Supplementaryinformationc.pdf Cite Share Download PDF Status: Published Journal Publication published 23 Dec, 2025 Read the published version in Cancer Immunology, Immunotherapy → Version 1 posted Editorial decision: Revision requested 08 Oct, 2025 Reviews received at journal 07 Oct, 2025 Reviews received at journal 16 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers invited by journal 03 Sep, 2025 Editor assigned by journal 30 Aug, 2025 Submission checks completed at journal 30 Aug, 2025 First submitted to journal 29 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7486809","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":510740294,"identity":"dec7ee0b-b25f-42f9-901b-4d28c4a824fc","order_by":0,"name":"Kamila Kaminska","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Kamila","middleName":"","lastName":"Kaminska","suffix":""},{"id":510740295,"identity":"11a21d0f-5adf-446e-9ebc-5adf2b9bf60e","order_by":1,"name":"Bengt Phung","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Bengt","middleName":"","lastName":"Phung","suffix":""},{"id":510740296,"identity":"7df6f2b1-b756-4db4-865d-0175f0dd0e79","order_by":2,"name":"Jacob Karlström","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Jacob","middleName":"","lastName":"Karlström","suffix":""},{"id":510740297,"identity":"f306b922-dd52-42d6-b412-33dc31f54313","order_by":3,"name":"Martin Lauss","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Lauss","suffix":""},{"id":510740298,"identity":"0b891ead-2b20-4e60-bdd5-bc7074a6739e","order_by":4,"name":"Katja Harbst","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Katja","middleName":"","lastName":"Harbst","suffix":""},{"id":510740299,"identity":"befb1ac0-ca25-44fb-a651-307b7dde6e51","order_by":5,"name":"Teresa Svensson","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Teresa","middleName":"","lastName":"Svensson","suffix":""},{"id":510740300,"identity":"06a4b302-bfc5-433c-8e46-00dda34c9d6e","order_by":6,"name":"Kristian Pietras","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Kristian","middleName":"","lastName":"Pietras","suffix":""},{"id":510740301,"identity":"29552e6e-e28b-4f91-83d1-fb74e0af23b8","order_by":7,"name":"Kari Nielsen","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Kari","middleName":"","lastName":"Nielsen","suffix":""},{"id":510740303,"identity":"d4da715e-54f7-4c6d-b475-249177b6eb84","order_by":8,"name":"Ana Carneiro","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"","lastName":"Carneiro","suffix":""},{"id":510740305,"identity":"9a223e85-bb13-4d95-8657-31dfa56de834","order_by":9,"name":"Henrik Ekedahl","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Henrik","middleName":"","lastName":"Ekedahl","suffix":""},{"id":510740307,"identity":"b7f72f8d-3889-4056-ae7b-a87864eec77a","order_by":10,"name":"Karolin Isaksson","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Karolin","middleName":"","lastName":"Isaksson","suffix":""},{"id":510740309,"identity":"283dd037-24d3-4a60-bfee-ef72cdd9e4c8","order_by":11,"name":"Göran Jönsson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCklEQVRIiWNgGAWjYPCCAwyMDRCWHBAbMDwgRYsxWEsCMVpgILGBkBZ5996HnwsY7jAwt/ce/vij4k762vbmDQyJbQz2/Di0GJ45biw9g+EZA2PPuTRpnjPPcredOVYA0pI4swGHlhlpDNI8DIcZGGfkmDEzth3O3XYjxwCkJcHgAE4tzL+hWow//mw7nG52/w1Yi709Di3yEmlsMFsMJHjbDieY3eABa2HcgMMvBjzH2Kx5GJ7xMPacMQP65bDhtjNpBQcSzkkkzsBlS3sb820ehjtyhu09xsAQOyxvdvzwxgcfymzs+XF4H+xHxn8MPIbICoCCEjicBbQFplIep5JRMApGwSgY8QAAyW1bIudygDoAAAAASUVORK5CYII=","orcid":"","institution":"Lund University","correspondingAuthor":true,"prefix":"","firstName":"Göran","middleName":"","lastName":"Jönsson","suffix":""}],"badges":[],"createdAt":"2025-08-29 08:53:47","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7486809/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7486809/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00262-025-04269-9","type":"published","date":"2025-12-23T15:57:53+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":90929266,"identity":"1083154d-116e-4b43-afde-0cd167e1d8a5","added_by":"auto","created_at":"2025-09-09 16:04:39","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":81837,"visible":true,"origin":"","legend":"\u003cp\u003ePatient-derive organotypic cultures from metastatic melanoma tissue. \u003cstrong\u003eA)\u003c/strong\u003e Overall scheme of the generation of organotypic cultures. \u003cstrong\u003eB)\u003c/strong\u003e Representative image of organoids growing in semi-solid media. Included are also representative hematoxylin and eosin staining images from four different organoids.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/e32452acec5c3635e5bd6d5f.jpg"},{"id":90929271,"identity":"322fb118-d1b0-437a-a0ba-8a26e9b1f92d","added_by":"auto","created_at":"2025-09-09 16:04:39","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":135910,"visible":true,"origin":"","legend":"\u003cp\u003eGenomic and transcriptomic properties of melanoma metastasis and matched organotypic culture. \u003cstrong\u003eA) \u003c/strong\u003ePrincipal component analysis of the 5000 most variable genes in metastatic tissue and matched semi-solid PDO and air-liquid PDO. \u003cstrong\u003eB)\u003c/strong\u003eHeatmap displaying gene expression values of the microenvironment- and cell populations-counter (MCP) scores. In addition, gene expression scores representing antigen presentation (HLA-A, HLA-B, HLA-C, B2M, TAP1), melanocytic (SOX10, MITF, TYR, PMEL) and dedifferentiated melanoma (SOX10, NGFR, AXL) are displayed. \u003cstrong\u003eC)\u003c/strong\u003e Boxplot of gene expression correlation values between matched metastatic lesion and PDO and values from unmatched metastatic lesion and PDO. \u003cstrong\u003eD)\u003c/strong\u003e Representative DNA copy number profiles generated from RNA sequencing data. Chromosomes 9 and 10 are highlighted since these are recurrently lost in melanoma tumors. Chromosomes 11 and 15 show differences between metastasis and PDO. \u003cstrong\u003eE)\u003c/strong\u003e Hotspot- and loss of function mutations in BRAF, NRAS and NF1 in matched metastasis and PDO. * marks patients with PDOs circled in the PCA plot in A).\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/6aaebc0231949f7d479c57f6.jpg"},{"id":90931570,"identity":"2fd1da4f-dc93-47ab-89a6-91b9bb58a8e4","added_by":"auto","created_at":"2025-09-09 16:20:39","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003eCellular and histological characterization of organoids. \u003cstrong\u003eA)\u003c/strong\u003e Representative flow cytometry analysis charts of organotypic cultures from patients 4, 6 and 13. Antibodies against CD3 and CD45 were used. \u003cstrong\u003eB)\u003c/strong\u003e Immunostaining of SOX10, CD8 and CD20 in organotypic cultures from Patients 4 and 6. \u003cstrong\u003eC)\u003c/strong\u003eImmunostaining of SOX10, CD8 and CD20 in organotypic culture and matched metastasis from patient 7. \u003cstrong\u003eD)\u003c/strong\u003e UMAP plot using single cell RNA sequencing data from organotypic culture and matched metastasis from Patient 6. \u003cstrong\u003eE)\u003c/strong\u003eBarplot of sample representation in each UMAP cluster. \u003cstrong\u003eF)\u003c/strong\u003e Dotplot showing expression average of melanoma related genes in each UMAP cluster. Average value of all cells belonging to metastasis and organoid separately are displayed.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/eada290edee10fe37b4185e7.png"},{"id":90930650,"identity":"f05ba3c3-1910-4e33-9c98-8fa5d57216d8","added_by":"auto","created_at":"2025-09-09 16:12:39","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003eT cell stimulation in organotypic cultures and clinical benefit of immune checkpoint blockade (ICB). \u003cstrong\u003eA)\u003c/strong\u003e Overall scheme of experimental setup. \u003cstrong\u003eB)\u003c/strong\u003e Microenvironment- and cell populations-counter (MCP) scores from baseline and PD1 therapy treated organotypic cultures. In addition, gene expression scores representing antigen presentation (HLA-A, HLA-B, HLA-C, B2M, TAP1), melanocytic (SOX10, MITF, TYR, PMEL) and dedifferentiated melanoma (SOX10, NGFR, AXL) are displayed. Cases are grouped according to patient ICB therapy response. \u003cstrong\u003eC) \u003c/strong\u003eBarplot showing number of specific T cell clones using T cell clonality analysis. Cases are grouped according to patient ICB therapy response. \u003cstrong\u003eD)\u003c/strong\u003e Representative immunostaining of three cases. Arrows highlight T cells in tumors. TLS refers to tertiary lymphoid structures. \u003cstrong\u003eE)\u003c/strong\u003e Spatial transcript profiling of two melanoma metastases from Patients 7 and 13, respectively. In Patient 7 arrows indicate antigen-presenting melanoma cells. In Patient 13 arrows indicate CD8\u003csup\u003e+\u003c/sup\u003e T cells. Zoom-in regions are marked by a square. \u003cstrong\u003eF)\u003c/strong\u003e Violin-plot of a composite gene expression score of checkpoint molecules (PDCD1, LAG3, TIGIT, HAVCR2) in CD8+ T cells from patients 13 (n=20) and 7 (n=261).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/e4df399ea5f9d852f131d851.png"},{"id":90929269,"identity":"29987e7d-d364-42bb-9505-f71992267b07","added_by":"auto","created_at":"2025-09-09 16:04:39","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003eICB induced transcriptional profiles in organotypic cultures. \u003cstrong\u003eA)\u003c/strong\u003e Heatmap of gene expression changes from baseline to PD1 blockade treated organotypic culture in cases from ICB responders (green bar) and ICB resistant cases (blue bar). Gene ontology analysis of each gene set is displayed. \u003cstrong\u003eB)\u003c/strong\u003e Heatmap of scaled mean expression of genes upregulated in organotypic cultures from patients with benefit of ICB. Single-cell RNA sequencing data were retrieved from Pozniak et al.\u003csup\u003e21\u003c/sup\u003e. The 14 genes exclusively expressed in T cells are indicated. \u003cstrong\u003eC)\u003c/strong\u003e Gene expression score of the 14 T cell specific genes was used on bulk RNA sequencing data of matched pre/on treatment samples (n=42) from Riaz et al.\u003csup\u003e25\u003c/sup\u003e. \u003cstrong\u003eD)\u003c/strong\u003e Kaplan-Meier progression-free survival (PFS) and overall survival (OS) analysis. Stratification is based on the 14-gene T cell specific signature score on bulk RNA sequencing data of metastatic melanoma patients (n=69) from Gide et al\u003csup\u003e26\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/f74d027a991dfafa1e005ba5.png"},{"id":99172883,"identity":"e06419ba-6e06-45ca-832d-510040ac0f9f","added_by":"auto","created_at":"2025-12-29 16:11:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":934793,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/697006f5-18a7-4503-87c1-eda2b2ac4d27.pdf"},{"id":90930652,"identity":"e4e1dbcf-497e-4045-851e-5419f52db25c","added_by":"auto","created_at":"2025-09-09 16:12:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1777409,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryinformationc.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7486809/v1/8e82d65703b1db0393456c4e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Modelling anti-tumor immune responses using patient-derived melanoma organoids","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMelanoma, a highly aggressive form of skin cancer that poses significant challenges in treatment, especially in its metastatic stage. However, during the last decade, treatment with clinical introduction of immunomodulatory therapies has dramatically changed the outcome of melanoma patients with advanced disease\u003csup\u003e1\u003c/sup\u003e. PD1 blocking antibodies are currently the first-line treatment in advanced melanoma given in the adjuvant or neoadjuvant setting. Despite the central role of PD1 blockade in the clinical management of cancer patients the functional consequences are not fully known. PD1 is expressed on activated T cells, B cells and dendritic cells\u003csup\u003e2-4\u003c/sup\u003e. The ligand of PD1, PDL1, is induced on melanoma cells in response to interferon gamma (IFNG), thereby activating a negative feedback loop that inhibits tumor-specific T cells\u003csup\u003e5\u003c/sup\u003e. Blocking PD1 reactivates these T cells allowing tumor-specific killing to occur. However, a deeper understanding of the molecular consequences of T cell activation within tumors is still needed. Recent studies describing the superior clinical efficacy of neoadjuvant ICB compared to adjuvant ICB underscore the need to understand the early immunological changes occurring within tumors\u003csup\u003e6\u003c/sup\u003e. Voabil et al.\u003csup\u003e7\u003c/sup\u003e used patient-derived tissue fragments (PDTF) to analyze early immunological consequences of PD1 blockade. Their analysis demonstrated that expression of several chemokines such as IFNG and CXCL10, in fragments from patients who responded clinically to ICB, irrespective of cancer type\u003csup\u003e7\u003c/sup\u003e. Alternative model systems to PDTFs include patient-derived organoids\u003csup\u003e8 9\u003c/sup\u003e. These models have the advantage of preserving the tumor microenvironment allowing the evaluation of immunomodulatory effects on both tumor and immune cells. Using patient-derived organoids Ou et al. showed that PD1 blockade activated CD8\u003csup\u003e+\u003c/sup\u003e T cells and induced tumor cell death confirming the utility of such models in cancer immunology research\u003csup\u003e10\u003c/sup\u003e. However, the use of such \u003cem\u003eex vivo\u003c/em\u003e models to develop new predictive biomarkers and exploring early transcriptional effects of PD1 blockade have not been fully elucidated.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn this study, we generated patient-derived organotypic cultures from fresh metastatic melanoma tissue. These cultures preserved several immune and molecular properties \u003cem\u003eex vivo\u003c/em\u003e. By stimulating T cells from both ICB responders and non-responders, we identified distinct transcriptional signatures that were predictive of patient outcomes following ICB therapy. Collectively, we show that patient-derived organotypic cultures is a valuable model for studying anti-tumor immune responses and discovering new predictive biomarkers. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003e\u003cem\u003ePatient samples and organotypic cultures\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTumor material used to establish patient derived organotypic cultures was acquired from patients undergoing surgical resection of metastatic melanoma at Skåne University Hospital, Kristianstad General Hospital and Helsingborg General Hospital included in BioMEL\u003csup\u003e11\u003c/sup\u003e (ClinicalTrials.gov ID NCT05446155). \u0026nbsp;All patients undergoing metastatic melanoma surgery (years 2021-2023) from whom tissue was obtained provided written informed consent for the collection of tissue and matched normal blood samples for research as approved by the local ethical board (Lund University Ethical Review Board, Dnr. 2013/101). The study adhered to the declaration of Helsinki. Tissue biopsies were processed fresh usually within 1-2 hours from the surgical procedure. Two types of patient-derived organotypic (PDO) cultures were set up from fresh tumor material Semi-solid PDO cultures modified from Vilgelm et al.\u003csup\u003e12\u003c/sup\u003e and air-liquid interface (ALI) cultures as in Neal et al.\u003csup\u003e13\u003c/sup\u003e with modifications (Supplementary information).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eT cell stimulation in patient derived organotypic cultures\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBoth types of cultures were treated for 1 week with 10 µg/ml nivolumab (Selleckchem) in the presence of 2 μg/ml anti-CD3 (clone HIT3a, BioLegend) and 2 μg/ml anti-CD28 (clone CD28.2, BioLegend) antibodies. Baseline and treated samples were collected for bulk RNA sequencing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGenomic analyses\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003ePDO cultures recovered from the matrix and original tumors were processed for RNA and DNA extraction with AllPrep DNA/RNA Mini Kit (Qiagen). Sequencing libraries were prepared using TruSeq Stranded mRNA Library Prep kit (Illumina), pooled and sequenced in a 2x150 bp paired-end setup (Illumina) on NovaSeq 6000 (Illimuna). Bulk RNA sequencing data were processed using HISAT and Stringtie\u003csup\u003e14\u003c/sup\u003e to obtain fragments per kilobase of transcript per million mapped reads (FPKM) values. Variants in BRAF, NRAS and NF1 were called from hisat bam files using VarDict v.1.8.2\u003csup\u003e15\u003c/sup\u003e. The resulting variants were annotated with Annovar \u003csup\u003e16\u003c/sup\u003e. Copy number alterations were inferred using InferCNV (version 1.18.1). An InfercnvObject was constructed from 71 bulk RNA-seq organoid samples pooled together with 15 bulk RNA-seq samples from benign nevi, which were designated as control samples. MiXCR\u003csup\u003e17\u003c/sup\u003e version 3.0.3 was applied to detect T cell receptor clones. VDJtools version 1.2.1 was run on MiXCR output\u003csup\u003e18\u003c/sup\u003e (Supplementary information).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eImmunostaining\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOrganoids and a fragment of original tumor tissue from each patient were fixed in 10% neutral buffered formalin solution (Sigma) for 24 hours and processed according to standard dehydration protocol and embedded in paraffin blocks. \u0026nbsp;Sections were taken and stained for SOX10 (BioCare Medical, clone BC34), CD20 (Roche, clone L26) and CD8 (Cell Marque Sigma-Aldrich, clone C8/144B). Immunoslides were scanned with NanoZoomer S60 (Hamamatsu) and processed with NDP.view2.8.24 (Hamamatsu Photonics K.K.) (Supplementary information).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFlow cytometry analysis\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003ePDO cultures were recovered from culture matrix as described above and further dissociated into single cell suspension. Antibodies used in cocktail in FACS buffer (1% FBS in DPBS): anti-CD45-APC-Vio®770 (clone 5B1; Miltenyi), anti-CD3-PE (clone BW264/56; Miltenyi) and anti-CD19-FITC (clone HIB19; BD). Samples were analyzed on FACSMelody Cell Sorter (BD Biosciences), data were analyzed with FlowJo v10.9.0.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSingle cell RNA sequencing\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe generated scRNAseq data using Chromium Next GEM Single Cell 3’ Kit (10x Genomics) according to manufacturer’s recommendations. Libraries were sequenced on NovaSeq6000 (Illumina) as per 10x Genomics User Guide. The h5 files were processed and merged using the R package Seurat 4.0.182, the data were reduced to protein-coding genes, translational (RPS/RPL) and mitochondrial genes (MT-), and genes which a maximum count \u0026lt; = 4 were removed. Cells with less than 500 expressed genes were removed. The data were normalized using SCTransform. Harmony embeddings were calculated from the principal components\u003csup\u003e19\u003c/sup\u003e. The data were visualized using UMAP and clusters were identified using Seurat (Supplementary information).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSingle cell spatial transcriptomics\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSingle cell spatial transcriptomic data were obtained using the CosMx platform (Nanostring, Seattle, WA) as described previously\u003csup\u003e20\u003c/sup\u003e. Stromal cells were distinguished as endothelial cells and fibroblasts. Tumor cells with B2M log-transformed expression \u0026gt;=3 were considered “MHC-I high”, tumor cells without any expression of MITF, TYR, MLANA and PMEL were considered “Melanin-low”, and else were considered “Melanin-high”.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExternal single cell and bulk RNA sequencing data\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSingle cell RNA sequencing data from Pozniak et al.\u003csup\u003e21\u003c/sup\u003e were downloaded as “Entire_TME.Rds” data file. The data were normalized using SCTransform\u003csup\u003e22\u003c/sup\u003e and initial zero counts were restored. PCA was performed and Harmony was used to adjust for 3‘ vs 5‘ library chemistry\u003csup\u003e19\u003c/sup\u003e. Clustering was performed using the Seurat (Seurat 4.3.0)\u003csup\u003e23\u003c/sup\u003e. UMAP was used to visualize the data and cluster identities were manually assigned using key marker gene expression. External bulk RNA sequencing data have been processed as described previously\u003csup\u003e24\u003c/sup\u003e. Data from Riaz et al.\u003csup\u003e25\u003c/sup\u003e and Gide et al.\u003csup\u003e26\u003c/sup\u003e were used (Supplementary information).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003ePatient-derived melanoma organotypic cultures\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, we aimed to develop model systems suitable for testing immunomodulatory drugs in melanoma. We obtained fresh tumor tissue at surgery from 31 patients diagnosed with metastatic melanoma. The cohort consisted of 17 lymph nodes, 11 cutaneous/subcutaneous metastases and three CNS metastases. We established patient-derived organotypic cultures from fresh tumor material with the aim of preserving a diverse tumor microenvironment. Two types of cultures were established: semi-solid patient-derived organotypic cultures\u003csup\u003e12\u003c/sup\u003e and air-liquid interface patient-derived organotypic cultures\u003csup\u003e13\u003c/sup\u003e (\u003cstrong\u003eFig.1a\u003c/strong\u003e). Semi-solid PDO cultures were considered successful upon the formation of three-dimensional spheroid structures within the first week of culture while successful air-liquid PDO cultures exhibited spheroids or dense tumor fragments that maintained their structure and integrity over time (\u003cstrong\u003eFig.1b\u003c/strong\u003e and \u003cstrong\u003eFig.S1\u003c/strong\u003e). The overall success rate was 71%, corresponding to successful establishment of material from 22 out of 31 samples using at least one of the two culture methods (\u003cstrong\u003eFig.1c\u003c/strong\u003e)\u003cem\u003e.\u0026nbsp;\u003c/em\u003eAir-liquid PDO cultures were successfully established in 11 out of 13 samples (85%), including some samples where semi-solid PDO cultures failed. In conclusion, we have successfully generated patient-derived organotypic cultures from different metastatic sites including, the brain, lymph node and subcutaneous skin tissues.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBaseline metastasis and baseline PDOs display genomic similarities\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNext, we generated whole transcriptomic data from tumor biopsy and matched patient-derived organotypic culture using RNA sequencing. Sufficient RNA was collected from 14 PDOs and matched metastatic deposit with different clinical characteristics (\u003cstrong\u003eTable 1\u003c/strong\u003e). \u0026nbsp;Using principal component analysis (PCA) we observed that metastasis and matched PDO were broadly similar in the transcriptional space (\u003cstrong\u003eFig.2a\u003c/strong\u003e). To identify cell-type specific differences between the experimental settings, we calculated microenvironment and cell populations-counter (MCP) scores. While most MCP signatures showed similar expression levels, monocytic, fibroblastic and endothelial cell signatures were elevated in baseline tumors compared to PDO (\u003cstrong\u003eFig.2b\u003c/strong\u003e). Reassuringly, T- and B cell signatures were preserved during organoid formation. Next, we aimed to statistically assess each PDO was more transcriptionally similar to its corresponding metastasis than to unrelated metastases. We calculated the correlation between each PDO and all analyzed metastatic tumor lesions. Matched PDO and metastases showed significantly higher correlation values compared to unmatched pairs, confirming that transcriptional properties are preserved during organotypic culturing (\u003cem\u003eP\u003c/em\u003e = 1.3x10\u003csup\u003e-8\u003c/sup\u003e, \u003cstrong\u003eFig.2c\u003c/strong\u003e). Next, we generated genome-wide copy number profiles using the RNA sequencing data. Frequent melanoma-specific copy number changes such as loss of chromosomes 9 and 10 were identified across all samples and those were largely preserved from metastatic tumors to the organotypic cultures. Minor differences were found, such as loss of chromosome 11 in the PDO in Patient 9. In the same case, we observed loss of chromosome 15 that was not detected in the PDO. Nevertheless, the transcriptional data were supported by matched PDO and metastases displaying an overall similar DNA copy number profiles (\u003cstrong\u003eFig.2d, Fig. S2\u003c/strong\u003e). This was further supported by somatic mutations in \u003cem\u003eBRAF\u003c/em\u003e, \u003cem\u003eNRAS\u003c/em\u003e and \u003cem\u003eNF1\u003c/em\u003e genes that were preserved during the formation of organoids (\u003cstrong\u003eFig.2e\u003c/strong\u003e). Moreover, a set of five semi-solid PDOs clustered together in the PCA plot (\u003cstrong\u003eFig.2a\u003c/strong\u003e). Detailed analysis of DNA copy number data and hot-spot mutations revealed devoid of genetic alterations. Histological analysis using SOX10 staining further confirmed the absence of melanoma cells (\u003cstrong\u003eFig.S2-3\u003c/strong\u003e). In all, these data suggest that these PDOs lack melanoma cells and thus do not represent their matched metastatic lesions. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn conclusion, the generation of PDOs preserves important genomic properties of the original metastases.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCellular and histological characteristics of PDOs\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNext, we evaluated cellular and histological characteristics of baseline PDO and their matched metastasis. First, we used flow cytometry analysis to determine the fraction of CD3\u003csup\u003e+\u003c/sup\u003e/CD45\u003csup\u003e+\u003c/sup\u003e T cells in the PDOs. The proportion of CD3\u003csup\u003e+\u003c/sup\u003e/CD45\u003csup\u003e+\u003c/sup\u003e T cells varied widely, ranging from 0.5% to 92%, reflecting the heterogeneity across cultures (\u003cstrong\u003eFig.3a\u003c/strong\u003e). In addition, different proportions of CD45\u003csup\u003e+\u003c/sup\u003e/CD3\u003csup\u003e-\u003c/sup\u003e immune cell populations were detected, suggesting the presence of immune cell types other than T cells. Secondly, we performed immunostaining of PDOs to determine immune cell localization and found a substantial heterogeneity in the presence and infiltration of CD8\u003csup\u003e+\u003c/sup\u003e T cells within the organoids (\u003cstrong\u003eFig.3b, Fig. S4\u003c/strong\u003e). Notably, CD8\u003csup\u003e+\u003c/sup\u003e T infiltration of the organoid was limited in the majority of PDOs. Visible CD20\u003csup\u003e+\u003c/sup\u003e B cells in conjunction with an organoid were only observed in two cases (\u003cstrong\u003eFig.3b\u003c/strong\u003e). \u0026nbsp;One case (Patient 7), despite massive infiltration of CD8\u003csup\u003e+\u003c/sup\u003e T and CD20\u003csup\u003e+\u003c/sup\u003e B cells within the organoids, SOX10\u003csup\u003e+\u003c/sup\u003e melanoma cells were still present (\u003cstrong\u003eFig.3c\u003c/strong\u003e). The matched metastatic lesion was subsequently analyzed by immunostaining for immune cells presence. Consistent with organoid findings, the metastatic lesion of organoid from Patient 7 displayed extensive CD8\u003csup\u003e+\u003c/sup\u003e T cell infiltration with prominent CD20\u003csup\u003e+\u003c/sup\u003e B cell clusters resembling tertiary lymphoid structures (TLS)\u003csup\u003e24\u003c/sup\u003e (\u003cstrong\u003eFig.3c\u003c/strong\u003e). In contrast, Patient 6 exhibited limited infiltration of CD8\u003csup\u003e+\u003c/sup\u003e T cells that was similarly reflected in the corresponding organoid (\u003cstrong\u003eFig.S5\u003c/strong\u003e). Clearly, not all T cells are infiltrating the actual organoid but are part of the organotypic culture as demonstrated in the flow cytometry data (\u003cstrong\u003eFig.3a\u003c/strong\u003e). Melanoma cell states are known to shift in response to microenvironmental cues\u003csup\u003e27\u003c/sup\u003e. To investigate this, we performed single cell RNA sequencing (scRNAseq) on the metastatic lesion from Patient 6 and, in parallel, on tumor organoids isolated from the corresponding culture. Melanoma cells were identified based on SOX10 expression and clustering with UMAP was used to define distinct melanoma cell states. Overall, 10.258 melanoma cells were used in the analysis. Previously, Pozniak et al. described melanoma cell states by scRNAseq\u003csup\u003e21\u003c/sup\u003e. To recapitulate these states, we used the geneset described in Pozniak et al. This approach identified five UMAP clusters with cluster 2 enriched for melanoma cells from the metastatic lesion and cluster 1 enriched for melanoma cells from the PDO (\u003cstrong\u003eFig.3d-e\u003c/strong\u003e). However, none of the clusters were exclusive to either metastatic lesion or PDO (\u003cstrong\u003eFig.3e\u003c/strong\u003e). Moreover, all clusters expressed both melanocytic and antigen-presenting molecules indicating only subtle changes between them and suggesting that overall melanoma cells from the PDO closely resemble those in the metastatic lesion (\u003cstrong\u003eFig.3f\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eIn conclusion, the presence of immune cells in baseline metastatic lesion is reflected in PDOs, suggesting an intact anti-tumor immune response. Moreover, melanoma cell states are preserved in the PDOs, reflecting those of the original tumors. \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTranscriptional effect of T cell stimulation is based on clinical response to ICB\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo test the relevance of PDO for evaluating anti-cancer immune responses, we used four PDOs derived from patients who experienced clinical benefit from PD1 blockade. As controls, we used four PDOs derived from patients who relapsed following either single-agent PD1 blockade or combined PD1/CTLA4 therapy. To determine the early effects of T cell stimulation, we treated PDOs with CD3 and CD28 agonists in combination with a PD1 blocker for 1 week. PDOs were processed for RNA sequencing pre-treatment and one week of treatment (\u003cstrong\u003eFig.4a\u003c/strong\u003e). We then applied MCP-counter to infer presence and abundance of cell types within the cultures. Broadly, baseline PDOs from patients having clinical benefit to ICB showed higher expression scores for antigen presentation, while those from ICB resistant metastases exhibited increased fibroblastic signatures. Although PDOs from patients with clinical benefit from ICB showed higher T cell-related gene expression scores, extensive heterogeneity was observed (\u003cstrong\u003eFig.4b\u003c/strong\u003e). Next, we determined T cell receptor clonality (TCR) and found a higher frequency of T cell clones in baseline PDOs from ICB responders. In three of these patients T cell stimulation led to clonal expansion, whereas no such expansion of was observed in cultures from non-responders (\u003cstrong\u003eFig.4c\u003c/strong\u003e). Moreover, PDOs from responders showed increased T cell diversity following T cell activation (\u003cstrong\u003eFig.S6\u003c/strong\u003e). Histology analyses of the matched metastatic lesions from patients 4, 7 and 12 identified distinct regions of immune clusters resembling tertiary lymphoid structures (\u003cstrong\u003eFig.4d\u003c/strong\u003e) suggesting that such melanomas with benefit of ICB are susceptible to T cell reinvigoration. Patients 4 and 12 both received adjuvant PD1 blockade and were recurrence-free after \u0026gt;1 year. Patient 7 in the cohort received adjuvant anti-PD1/LAG3 therapy, which was halted due to toxicity. A recurrence was detected after almost a year after surgery. The patient subsequently received anti-PD1/CTLA4 therapy, which was discontinued after two cycles. At the most recent follow-up (\u0026lt; 6months from treatment pause), no disease progression had been observed. In contrast, histological analysis of metastatic lesion from patient 13 displayed prominent CD8\u003csup\u003e+\u003c/sup\u003e T cell presence however spatially localized in clusters (\u003cstrong\u003eFig.4d\u003c/strong\u003e). The other three ICB resistant cases all had very few infiltrating CD8\u003csup\u003e+\u003c/sup\u003e T cells (\u003cstrong\u003eFig.S7\u003c/strong\u003e). Next, we wanted to understand why there was a T cell expansion in patients with clinical benefit of ICB while this was not observed in samples from patients resistant to ICB. Thus, we leveraged single cell spatial trancriptomics of two metastatic lesions from Patients 7 and 13. In Patient 7 we observed a wide range of different immune cells including activated CD8\u003csup\u003e+\u003c/sup\u003e T cells, cells of the monocytic lineage, B cells and other T cells (\u003cstrong\u003eFig.4e\u003c/strong\u003e). We also found a significant proportion of melanoma cells expressing MHC class I suggesting that they are actively presenting antigens. Such cells were frequently located in proximity to immune cells while melanoma cells spatially located distant from immune cells had a decreased expression of MHC class I (\u003cstrong\u003eFig.4e\u003c/strong\u003e). In contrast, we generated the same type of data from Patient 13 that had relapsed on adjuvant anti-PD1 therapy but still had a substantial number of T cells which did not show clonal expansion \u003cem\u003eex vivo\u003c/em\u003e (\u003cstrong\u003eFig.4c\u003c/strong\u003e). We found that this melanoma lacked tumor cells expressing MHC class I molecules and phenotypically such melanoma cells also had decreased melanocytic transcriptional properties. We also observed a massive infiltration of macrophages but also CD8\u003csup\u003e+\u003c/sup\u003e T cells (\u003cstrong\u003eFig.4e\u003c/strong\u003e). When analysing the transcriptional phenotypes of CD8\u003csup\u003e+\u003c/sup\u003e T cells in these two melanomas we found that CD8\u003csup\u003e+\u003c/sup\u003e T cells in Patient 7, who had benefit of ICB therapy, had expression of checkpoint molecules such as PD1, LAG3, TIM3 and TIGIT, while CD8\u003csup\u003e+\u003c/sup\u003e T cells from Patient 13 completely lacked expression of such checkpoint/activation molecules (P = 0.0007, \u003cstrong\u003eFig.4f\u003c/strong\u003e). \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCollectively, these findings indicate that T cells from patients who benefit of ICB are capable of expanding \u003cem\u003eex vivo\u003c/em\u003e and display transcriptional profiles consistent with activation and checkpoint engagement. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eT cell stimulation in PDOs from therapy naïve melanoma tissue confers a distinct transcriptional profile predictive of ICB clinical response\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe next determined the gene programs upregulated by T cell stimulation \u003cem\u003eex vivo\u003c/em\u003e in PDOs from patients with clinical benefit from ICB. Intriguingly, we observed strong induction of genes related to T cell activation including, CCL1, OX40, ICOS, CTLA4 and CXCL13 (\u003cstrong\u003eFig.5a\u003c/strong\u003e). Consistent with the T cell clonality results, Patients 4 and 7 displayed the strongest effect on gene expression changes. However, similar transcriptional responses were also observed in the other two PDOs from patients who had benefited from ICB. In these ICB responder cases, gene ontology analysis revealed enrichment of general-immune system processes such as lymphocyte and T cell activation were enriched. Intriguingly, these genes remained unaltered in T cell-stimulated PDOs derived from patients who previously had relapsed on ICB (\u003cstrong\u003eFig.5a\u003c/strong\u003e). Next, we investigated genes upregulated in PDOs from patients who had relapsed on ICB including Patient 13 whose baseline culture contained a substantial number of T cells (\u003cstrong\u003eFig.4c\u003c/strong\u003e). The genes displaying the highest fold change between baseline and stimulated PDOs from ICB-resistant tissue were enriched for innate immune response genes. Notably, this transcriptional increase was predominantly found in a single patient, Patient 13. This patient relapsed 9 months after adjuvant anti-PD1 therapy and underwent surgical resection of a subcutaneous lesion prior to initiating dual PD1/CTLA4 blockade. At eight-month follow-up, the patient remained recurrence-free.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGiven the consistent upregulation of a transcriptional program in PDOs treated with T cell stimulatory molecules derived from patients with clinical benefit from ICB, we wanted to explore this in detail. Analysis of single cell RNA sequencing data from melanoma metastases\u003csup\u003e21\u003c/sup\u003e revealed that 14 of the upregulated genes were exclusively expressed in T cells (\u003cstrong\u003eFig.5b\u003c/strong\u003e). To determine the predictive value of these 14 T cell specific genes, we created a composite score and applied this on bulk RNA sequencing data from two independent datasets\u003csup\u003e25 26\u003c/sup\u003e. In the Riaz et al.\u003csup\u003e25\u003c/sup\u003e RNA sequencing data on pre- and on-treatment biopsies from 42 metastatic melanoma patients were available. Confirming our \u003cem\u003eex vivo\u003c/em\u003e results, we found that the 14-gene T cell signature was increased in on-treatment biopsies from patients with stable disease, partial response or complete response (\u003cstrong\u003eFig.5c\u003c/strong\u003e). Such increase in gene expression signature was not observed in patients with progressive disease on ICB (\u003cstrong\u003eFig.5c\u003c/strong\u003e). \u0026nbsp;Finally, we used RNA sequencing data from 69 melanoma patients receiving anti-PD1 or anti-PD1/CTLA4 therapy\u003csup\u003e26\u003c/sup\u003e and applied the T cell signature on these samples. Stratifying patients into tertiles based on the T cell gene expression score displayed significant difference in progression-free and overall survival (\u003cstrong\u003eFig.5d\u003c/strong\u003e, \u003cem\u003eP\u003c/em\u003e = 0.01) in patients with highest T cell gene expression score.\u003c/p\u003e\n\u003cp\u003eIn conclusion, transcriptomic signature derived from \u003cem\u003eex vivo\u0026nbsp;\u003c/em\u003emelanoma models can be utilized to predict ICB clinical response.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we generated and validated PDO systems from metastatic melanoma. These models serve as physiologically relevant models to evaluate immunomodulatory therapies. Utilizing fresh tumor tissues from 31 patients, we tested two culture approaches designed to preserve the cellular heterogeneity and spatial organization of the tumor microenvironment (TME). The overall success rate was 71%, with the air-liquid PDOs showing a higher establishment rate (85%). Notably, the latter succeeded even in cases where semi-solid PDOs failed, indicating the robustness of this model for capturing TME complexity \u003cem\u003eex vivo\u003c/em\u003e. These results corroborate previous efforts to develop organotypic culture systems for human cancers\u003csup\u003e13\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn our study, transcriptomic profiles of the PDOs closely matched their parental metastatic lesions. Although microenvironment-associated gene signatures such as monocytic, fibroblastic, and endothelial components were decreased in PDOs, critical immune cell lineages, including T and B cells, were preserved. These findings are consistent with the notion that organoid cultures can partially retain the TME, especially lymphoid components, which play pivotal roles in mediating response to immunotherapy\u003csup\u003e28\u003c/sup\u003e. Furthermore, matched tumor-PDO comparisons revealed significantly higher transcriptional correlation than unmatched pairs, confirming that PDO largely conserves tumor-intrinsic transcriptional programs. Copy number variation analysis from RNA-seq data also revealed conserved melanoma-specific alterations, including frequent losses of chromosomes 9 and 10, which are well-documented in melanoma pathogenesis\u003csup\u003e29\u003c/sup\u003e. Single-cell RNA sequencing (scRNA-seq) of melanoma cells from matched tumor and PDO identified overlapping transcriptional states, characterized by shared expression of melanocytic markers and antigen-presentation genes. This suggests that key tumor-intrinsic programs are preserved \u003cem\u003eex vivo\u003c/em\u003e. Further confirming that molecular states are preserved during PDO formation immunostaining revealed variable infiltration of CD8\u003csup\u003e+\u003c/sup\u003e T cells in the organoids. In one case (Patient 7), substantial infiltration was observed and reflected the matched metastatic lesion. In contrast, Patient 6 exhibited minimal immune infiltration in both the PDO and tumor. These findings support the reliability of PDOs in reflecting the immune landscape of the original tissue, consistent with previous reports in melanoma organoids\u003csup\u003e10\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eTo investigate the utility of PDOs in predicting response to ICB therapies, we stimulated organotypic cultures from patients with known clinical benefit from PD1-based therapy, as well as from patients who had relapsed after ICB, using T cell agonists and PD1 blockade. Post-treatment transcriptional data revealed significant upregulation of T cell activation signatures (e.g., CCL1, OX40, ICOS, CTLA4, CXCL13) in PDOs from ICB responders. In contrast, these changes were absent in PDOs derived from ICB-resistant cases. The chemokine CXCL13, for instance, is expressed by activated T cells and other stromal cells and plays a central role to support TLS formation. Previous work has shown its predictive value for ICB responses in melanoma\u003csup\u003e30\u003c/sup\u003e. Clonotype analysis revealed T cell expansion and increased TCR diversity in responder PDOs, but not in resistant cases, suggesting that the \u003cem\u003eex vivo\u003c/em\u003e models recapitulate \u003cem\u003ein vivo\u003c/em\u003e immune responsiveness. Indeed, the PDOs showing the strongest T cell expansion following PD1 blockade were derived from matched metastatic lesions harboring tertiary lymphoid structures (TLS). TLS are linked to improved outcomes in melanoma and other cancers due to their role in sustaining localized anti-tumor immune responses\u003csup\u003e24 31\u003c/sup\u003e. To explore the mechanism underlying these observations, we performed single cell spatial transcriptomics on matched tumors samples. In ICB responders, melanoma cells expressing MHC class I were found in proximity to activated CD8\u003csup\u003e+\u003c/sup\u003e T cells, consistent with effective antigen presentation and T cell priming. Conversely, in ICB-resistant cases, such as Patient 13, melanoma cells lacked MHC class I expression and melanocytic markers and CD8\u003csup\u003e+\u003c/sup\u003e T cells failed to express activation or checkpoint markers (e.g., PD-1, LAG3, TIM3), suggesting a dysfunctional immune microenvironment. These observations are supported by earlier studies in that have linked MHC class I loss and antigen presentation defects with ICB resistance\u003csup\u003e32-34\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUsing the generated bulk RNA sequencing data from stimulated PDOs of ICB responders we derived a 14-gene T cell activation signature. This signature was validated in independent clinical datasets\u003csup\u003e25 26\u003c/sup\u003e. These findings suggest that PDOs, when used in conjunction with T cell stimulation assays, can serve as a functional readout of a patient’s capacity to mount an immune response and may aid in stratifying patients for immunotherapy.\u003c/p\u003e\n\u003cp\u003eIn conclusion, our findings establish that patient-derived organotypic cultures from metastatic melanoma retain key genomic, cellular, and immunological features of the original tumor. The ability to simulate T cell activation and monitor clonal expansion and transcriptional changes \u003cem\u003eex vivo,\u003c/em\u003e provides a promising avenue for biomarker discovery and personalized immunotherapy testing.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank research nurses Marie Sj\u0026ouml;gren, Kerstin Reistad and Carina Eriksson for assisting in tissue handling. We thank Jari H\u0026auml;kkinen for assistance on aligning the RNA sequencing data. We thank Joanna Pozniak at KU Leuven for access to the single cell RNA sequencing data from patients receiving immune checkpoint blockade. We also thank the clinicians at Helsingborg Hospital, Kristianstad Hospital, and Sk\u0026aring;ne University Hospital for including patients in the BioMEL study. The authors would like to acknowledge Clinical Genomics Lund, SciLifeLab and Center for Translational Genomics (CTG), Lund University, for providing expertise and service with sequencing and analysis. The computations and data handling were enabled by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS) at Uppsala Multidisciplinary Center for Advanced Computational Science (UPPMAX) partially funded by the Swedish Research Council through grant agreement no. 2018-05973. We also thank Nanostring for CosMx experimental analyses.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData availability\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eRequests for information and resources should be directed to and will be fulfilled by the lead contact, G\u0026ouml;ran J\u0026ouml;nsson ([email protected]). Raw sequencing data are regarded as personal information, and by Swedish law, they cannot be made publicly accessible. However, information and mechanisms for data access can be obtained by contacting the corresponding author.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Swedish Research Council (Vetenskapsr\u0026aring;det Dnr 2018-02786, Dnr 2022-00871, GJ), Swedish Cancer Society (19 0458 Pj, 22 2105 Pj, GJ), Berta Kamprad Foundation (GJ) and the governmental funding for healthcare research (ALF, GJ and KN), Knut and Alice Wallenberg Foundation (KAW 2022.0066, GJ) and G\u0026ouml;ran Gustafsson Foundation (GJ). KN was supported by the governmental funding for healthcare research (ALF), the S.R. Gorthon foundation, Hudfonden/Welander-Finsen foundation, Research grants of the Southern health care region, the Gyllenstiernska Krapperup foundation and by the Inga and John Hain foundation for medical research. We gratefully acknowledge the support by Mrs. Berta Kamprad\u0026apos;s Cancer Foundation to the L2CancerBridge program at CREATE Health Cancer Center.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eK.K. conducted single cell sequencing, immunostaining, data analysis, design of study and writing the manuscript. B.P. conducted all immunostaining and analysis of such data. K.H. conducted RNA-, TCR and single cell RNA sequencing. K.N., T.S., A.C., H.E and K.I. collected clinical information on patients and set up protocols for collection of viable tumor tissue from patients with melanoma. K.P conducted flow cytometry analysis. M.L and J.K conducted data analysis on bulk and single cell RNA sequencing. G.J. supervised and designed study and wrote the manuscript. All authors have read and approved the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWolchok JD, Chiarion-Sileni V, Rutkowski P, et al. Final, 10-Year Outcomes with Nivolumab plus Ipilimumab in Advanced Melanoma. \u003cem\u003eN Engl J Med\u003c/em\u003e 2025;392(1):11-22. doi: 10.1056/NEJMoa2407417 [published Online First: 20240915]\u003c/li\u003e\n\u003cli\u003evan der Leun AM, Thommen DS, Schumacher TN. CD8(+) T cell states in human cancer: insights from single-cell analysis. \u003cem\u003eNat Rev Cancer\u003c/em\u003e 2020;20(4):218-32. doi: 10.1038/s41568-019-0235-4 [published Online First: 20200205]\u003c/li\u003e\n\u003cli\u003eXiao X, Lao XM, Chen MM, et al. PD-1hi Identifies a Novel Regulatory B-cell Population in Human Hepatoma That Promotes Disease Progression. \u003cem\u003eCancer Discov\u003c/em\u003e 2016;6(5):546-59. doi: 10.1158/2159-8290.CD-15-1408 [published Online First: 20160229]\u003c/li\u003e\n\u003cli\u003eKrempski J, Karyampudi L, Behrens MD, et al. Tumor-infiltrating programmed death receptor-1+ dendritic cells mediate immune suppression in ovarian cancer. \u003cem\u003eJ Immunol\u003c/em\u003e 2011;186(12):6905-13. doi: 10.4049/jimmunol.1100274 [published Online First: 20110506]\u003c/li\u003e\n\u003cli\u003eGarcia-Diaz A, Shin DS, Moreno BH, et al. Interferon Receptor Signaling Pathways Regulating PD-L1 and PD-L2 Expression. \u003cem\u003eCell Rep\u003c/em\u003e 2017;19(6):1189-201. doi: 10.1016/j.celrep.2017.04.031\u003c/li\u003e\n\u003cli\u003eBlank CU, Lucas MW, Scolyer RA, et al. Neoadjuvant Nivolumab and Ipilimumab in Resectable Stage III Melanoma. \u003cem\u003eN Engl J Med\u003c/em\u003e 2024;391(18):1696-708. doi: 10.1056/NEJMoa2402604 [published Online First: 20240602]\u003c/li\u003e\n\u003cli\u003eVoabil P, de Bruijn M, Roelofsen LM, et al. An ex vivo tumor fragment platform to dissect response to PD-1 blockade in cancer. \u003cem\u003eNat Med\u003c/em\u003e 2021;27(7):1250-61. doi: 10.1038/s41591-021-01398-3 [published Online First: 2021/07/10]\u003c/li\u003e\n\u003cli\u003eJenkins RW, Aref AR, Lizotte PH, et al. Ex Vivo Profiling of PD-1 Blockade Using Organotypic Tumor Spheroids. \u003cem\u003eCancer Discov\u003c/em\u003e 2018;8(2):196-215. doi: 10.1158/2159-8290.CD-17-0833 [published Online First: 2017/11/05]\u003c/li\u003e\n\u003cli\u003eSun L, Kang X, Ju H, et al. A human mucosal melanoma organoid platform for modeling tumor heterogeneity and exploring immunotherapy combination options. \u003cem\u003eSci Adv\u003c/em\u003e 2023;9(43):eadg6686. doi: 10.1126/sciadv.adg6686 [published Online First: 20231027]\u003c/li\u003e\n\u003cli\u003eOu L, Liu S, Wang H, et al. Patient-derived melanoma organoid models facilitate the assessment of immunotherapies. \u003cem\u003eEBioMedicine\u003c/em\u003e 2023;92:104614. doi: 10.1016/j.ebiom.2023.104614 [published Online First: 20230523]\u003c/li\u003e\n\u003cli\u003eHelkkula T, Christensen G, Ingvar C, et al. BioMEL: a translational research biobank of melanocytic lesions and melanoma. \u003cem\u003eBMJ Open\u003c/em\u003e 2024;14(2):e069694. doi: 10.1136/bmjopen-2022-069694 [published Online First: 20240202]\u003c/li\u003e\n\u003cli\u003eVilgelm AE, Bergdorf K, Wolf M, et al. Fine-Needle Aspiration-Based Patient-Derived Cancer Organoids. \u003cem\u003eiScience\u003c/em\u003e 2020;23(8):101408. doi: 10.1016/j.isci.2020.101408 [published Online First: 20200724]\u003c/li\u003e\n\u003cli\u003eNeal JT, Li X, Zhu J, et al. Organoid Modeling of the Tumor Immune Microenvironment. \u003cem\u003eCell\u003c/em\u003e 2018;175(7):1972-88 e16. doi: 10.1016/j.cell.2018.11.021 [published Online First: 2018/12/15]\u003c/li\u003e\n\u003cli\u003ePertea M, Kim D, Pertea GM, et al. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. \u003cem\u003eNat Protoc\u003c/em\u003e 2016;11(9):1650-67. doi: 10.1038/nprot.2016.095 [published Online First: 2016/08/26]\u003c/li\u003e\n\u003cli\u003eLai Z, Markovets A, Ahdesmaki M, et al. VarDict: a novel and versatile variant caller for next-generation sequencing in cancer research. \u003cem\u003eNucleic Acids Res\u003c/em\u003e 2016;44(11):e108. doi: 10.1093/nar/gkw227 [published Online First: 20160407]\u003c/li\u003e\n\u003cli\u003eWang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. \u003cem\u003eNucleic Acids Res\u003c/em\u003e 2010;38(16):e164. doi: 10.1093/nar/gkq603 [published Online First: 2010/07/06]\u003c/li\u003e\n\u003cli\u003eBolotin DA, Poslavsky S, Mitrophanov I, et al. MiXCR: software for comprehensive adaptive immunity profiling. \u003cem\u003eNat Methods\u003c/em\u003e 2015;12(5):380-1. doi: 10.1038/nmeth.3364\u003c/li\u003e\n\u003cli\u003eShugay M, Bagaev DV, Turchaninova MA, et al. VDJtools: Unifying Post-analysis of T Cell Receptor Repertoires. \u003cem\u003ePLoS Comput Biol\u003c/em\u003e 2015;11(11):e1004503. doi: 10.1371/journal.pcbi.1004503 [published Online First: 20151125]\u003c/li\u003e\n\u003cli\u003eKorsunsky I, Millard N, Fan J, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. \u003cem\u003eNat Methods\u003c/em\u003e 2019;16(12):1289-96. doi: 10.1038/s41592-019-0619-0 [published Online First: 20191118]\u003c/li\u003e\n\u003cli\u003eElewaut A, Estivill G, Bayerl F, et al. Cancer cells impair monocyte-mediated T cell stimulation to evade immunity. \u003cem\u003eNature\u003c/em\u003e 2025;637(8046):716-25. doi: 10.1038/s41586-024-08257-4 [published Online First: 20241127]\u003c/li\u003e\n\u003cli\u003ePozniak J, Pedri D, Landeloos E, et al. A TCF4-dependent gene regulatory network confers resistance to immunotherapy in melanoma. \u003cem\u003eCell\u003c/em\u003e 2024;187(1):166-83 e25. doi: 10.1016/j.cell.2023.11.037\u003c/li\u003e\n\u003cli\u003eHafemeister C, Satija R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression. \u003cem\u003eGenome Biol\u003c/em\u003e 2019;20(1):296. doi: 10.1186/s13059-019-1874-1 [published Online First: 20191223]\u003c/li\u003e\n\u003cli\u003eButler A, Hoffman P, Smibert P, et al. Integrating single-cell transcriptomic data across different conditions, technologies, and species. \u003cem\u003eNat Biotechnol\u003c/em\u003e 2018;36(5):411-20. doi: 10.1038/nbt.4096 [published Online First: 20180402]\u003c/li\u003e\n\u003cli\u003eCabrita R, Lauss M, Sanna A, et al. Tertiary lymphoid structures improve immunotherapy and survival in melanoma. \u003cem\u003eNature\u003c/em\u003e 2020;577(7791):561-65. doi: 10.1038/s41586-019-1914-8 [published Online First: 2020/01/17]\u003c/li\u003e\n\u003cli\u003eRiaz N, Havel JJ, Makarov V, et al. Tumor and Microenvironment Evolution during Immunotherapy with Nivolumab. \u003cem\u003eCell\u003c/em\u003e 2017;171(4):934-49 e15. doi: 10.1016/j.cell.2017.09.028\u003c/li\u003e\n\u003cli\u003eGide TN, Quek C, Menzies AM, et al. Distinct Immune Cell Populations Define Response to Anti-PD-1 Monotherapy and Anti-PD-1/Anti-CTLA-4 Combined Therapy. \u003cem\u003eCancer Cell\u003c/em\u003e 2019;35(2):238-55 e6. doi: 10.1016/j.ccell.2019.01.003 [published Online First: 2019/02/13]\u003c/li\u003e\n\u003cli\u003eKarras P, Bordeu I, Pozniak J, et al. A cellular hierarchy in melanoma uncouples growth and metastasis. \u003cem\u003eNature\u003c/em\u003e 2022;610(7930):190-98. doi: 10.1038/s41586-022-05242-7 [published Online First: 20220921]\u003c/li\u003e\n\u003cli\u003eDijkstra KK, Cattaneo CM, Weeber F, et al. Generation of Tumor-Reactive T Cells by Co-culture of Peripheral Blood Lymphocytes and Tumor Organoids. \u003cem\u003eCell\u003c/em\u003e 2018;174(6):1586-98 e12. doi: 10.1016/j.cell.2018.07.009 [published Online First: 20180809]\u003c/li\u003e\n\u003cli\u003eTCGA. Genomic Classification of Cutaneous Melanoma. \u003cem\u003eCell\u003c/em\u003e 2015;161(7):1681-96. doi: 10.1016/j.cell.2015.05.044 [published Online First: 2015/06/20]\u003c/li\u003e\n\u003cli\u003eBecht E, Giraldo NA, Lacroix L, et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. \u003cem\u003eGenome Biol\u003c/em\u003e 2016;17(1):218. doi: 10.1186/s13059-016-1070-5\u003c/li\u003e\n\u003cli\u003eHelmink BA, Reddy SM, Gao J, et al. B cells and tertiary lymphoid structures promote immunotherapy response. \u003cem\u003eNature\u003c/em\u003e 2020;577(7791):549-55. doi: 10.1038/s41586-019-1922-8 [published Online First: 2020/01/17]\u003c/li\u003e\n\u003cli\u003eZaretsky JM, Garcia-Diaz A, Shin DS, et al. Mutations Associated with Acquired Resistance to PD-1 Blockade in Melanoma. \u003cem\u003eN Engl J Med\u003c/em\u003e 2016;375(9):819-29. doi: 10.1056/NEJMoa1604958 [published Online First: 2016/07/20]\u003c/li\u003e\n\u003cli\u003eSade-Feldman M, Jiao YJ, Chen JH, et al. Resistance to checkpoint blockade therapy through inactivation of antigen presentation. \u003cem\u003eNat Commun\u003c/em\u003e 2017;8(1):1136. doi: 10.1038/s41467-017-01062-w\u003c/li\u003e\n\u003cli\u003eLauss M, Phung B, Borch TH, et al. Molecular patterns of resistance to immune checkpoint blockade in melanoma. \u003cem\u003eNat Commun\u003c/em\u003e 2024;15(1):3075. doi: 10.1038/s41467-024-47425-y [published Online First: 20240409]\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 1","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eClinical features of samples where organotypic cultures were included in the transcriptomic analyses. ICB \u0026ndash; immune checkpoint blockade\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"354\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 246px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eClinical feature\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eFraction\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003eGender\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eMale \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003cp\u003eFemale \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=11, 79%\u003c/p\u003e\n \u003cp\u003eN=3, 21%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003eBRAF mutation\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e(clinical testing)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eV600E \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eWt \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eNA \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=4, 29%\u003c/p\u003e\n \u003cp\u003eN=8, 57%\u003c/p\u003e\n \u003cp\u003eN=2, 14%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003ePrimary tumor site\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eLower extremity\u003c/p\u003e\n \u003cp\u003eUpper extremity\u003c/p\u003e\n \u003cp\u003eTrunk\u003c/p\u003e\n \u003cp\u003eHN\u003c/p\u003e\n \u003cp\u003eUnknown primary\u003c/p\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=2, 14%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003cp\u003eN=4, 29%\u003c/p\u003e\n \u003cp\u003eN=5, 36%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003ePrimary tumor type\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eSSM\u003c/p\u003e\n \u003cp\u003eNM\u003c/p\u003e\n \u003cp\u003eMucosal\u003c/p\u003e\n \u003cp\u003eUnknown primary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=7, 50%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003cp\u003eN=5, 36%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003ePrimary tumor Breslow\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003e5.8 mm,\u003c/p\u003e\n \u003cp\u003erange 2.2-14 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003eMetastasis\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eLN\u003c/p\u003e\n \u003cp\u003eSC\u003c/p\u003e\n \u003cp\u003eCNS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=10, 71%\u003c/p\u003e\n \u003cp\u003eN=3, 21%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003eStage at metastasis\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003eIIIB\u003c/p\u003e\n \u003cp\u003eIIIC\u003c/p\u003e\n \u003cp\u003eIIID\u003c/p\u003e\n \u003cp\u003eM1a\u003c/p\u003e\n \u003cp\u003eM1d\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=3, 21%\u003c/p\u003e\n \u003cp\u003eN=6, 43%\u003c/p\u003e\n \u003cp\u003eN=2, 14%\u003c/p\u003e\n \u003cp\u003eN=2, 14%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cem\u003eSampling with regards to treatment\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003ePre-ICB treatment\u003c/p\u003e\n \u003cp\u003eRelapse after ICB\u003c/p\u003e\n \u003cp\u003eOn-ICB treatment\u003c/p\u003e\n \u003cp\u003eTherapy na\u0026iuml;ve\u003c/p\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003eN=6, 43%\u003c/p\u003e\n \u003cp\u003eN=5, 36%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003cp\u003eN=1, 7%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cancer-immunology-immunotherapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ciim","sideBox":"Learn more about [Cancer Immunology, Immunotherapy](http://link.springer.com/journal/262)","snPcode":"262","submissionUrl":"https://submission.nature.com/new-submission/262/3","title":"Cancer Immunology, Immunotherapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Cancer, melanoma, immune response, patient-derived organoid, PD1","lastPublishedDoi":"10.21203/rs.3.rs-7486809/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7486809/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Immune checkpoint blockade (ICB) therapy can restore T cell function in tumors, but not all patients benefit, and the mechanisms behind this remain unclear. In this study, we used patient-derived organotypic (PDO) cultures from metastatic melanoma to examine transcriptomic and cellular changes following ex vivo T cell stimulation. Genomic and transcriptomic features were preserved during PDO formation, capturing melanoma heterogeneity. PDOs from ICB-responsive patients showed rapid T cell expansion upon T cell stimulation, unlike those from ICB-resistant tissue. Resistant tissue harbored T cells lacking activation and checkpoint markers, suggesting non-tumor-reactive T cells. A T cell-specific transcriptomic score, activated in responsive PDOs, correlated with improved overall and relapse-free survival in metastatic melanoma patients treated with ICB. These findings demonstrate that ex vivo analysis is a viable tool to investigate mechanisms of ICB response and may help identify predictive biomarkers for patient outcome.","manuscriptTitle":"Modelling anti-tumor immune responses using patient-derived melanoma organoids","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-09 16:04:35","doi":"10.21203/rs.3.rs-7486809/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-08T13:41:22+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-07T16:12:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-16T23:17:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"192044681261197293132773907987233868208","date":"2025-09-03T20:09:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"243959150937096948189101862933929190288","date":"2025-09-03T13:10:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"16815367619633522004558812515158669304","date":"2025-09-03T11:00:28+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-03T07:24:44+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-30T05:52:55+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-30T05:52:36+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cancer Immunology, Immunotherapy","date":"2025-08-29T08:43:24+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cancer-immunology-immunotherapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ciim","sideBox":"Learn more about [Cancer Immunology, Immunotherapy](http://link.springer.com/journal/262)","snPcode":"262","submissionUrl":"https://submission.nature.com/new-submission/262/3","title":"Cancer Immunology, Immunotherapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"4bafea59-f307-4848-aaf5-ae6131b1ddd7","owner":[],"postedDate":"September 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-29T16:09:56+00:00","versionOfRecord":{"articleIdentity":"rs-7486809","link":"https://doi.org/10.1007/s00262-025-04269-9","journal":{"identity":"cancer-immunology-immunotherapy","isVorOnly":false,"title":"Cancer Immunology, Immunotherapy"},"publishedOn":"2025-12-23 15:57:53","publishedOnDateReadable":"December 23rd, 2025"},"versionCreatedAt":"2025-09-09 16:04:35","video":"","vorDoi":"10.1007/s00262-025-04269-9","vorDoiUrl":"https://doi.org/10.1007/s00262-025-04269-9","workflowStages":[]},"version":"v1","identity":"rs-7486809","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7486809","identity":"rs-7486809","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}