Low success rates in clinical drug development can be largely attributed to the poor predictive power of existing preclinical models. Microphysiological systems (MPS) have greatly advanced in vitro modeling; however, current platforms do not adequately support long-term sampling and often fail to recapitulate nutrient and drug exposure dynamics. To address these limitations, we established a machine vision-guided MPS with real-time fluidic control that enables fully automated periodic sampling with high temporal resolution, media replenishment, and programmable dosing, allowing for the simulation of dynamic nutritional or pharmacological exposure scenarios. We showcase the system's capability by emulating physiological insulin profiles and repeated-dose pharmacokinetic exposures over multiple weeks. Furthermore, pharmacokinetically accurate acetaminophen exposure in 3D primary human liver spheroids mimicking an acute overdose rapidly induced liver toxicity, as evidenced by aminotransferase release, cytokine secretion and a drop in cellular ATP. In contrast, dose-equivalent constant exposure patterns did not elicit detectable hepatotoxicity. Mechanistically, targeted proteomics of sampled supernatants and Cell Painting revealed that toxicity was paralleled by disrupted lipid homeostasis, loss of tight junctions and extracellular matrix remodeling. These results demonstrate the robustness and versatility of the machine vision-guided automated microphysiological platform and underscore the importance of incorporating drug exposure dynamics for mechanistic toxicology.

Abstract Lung cancer is the most frequently diagnosed malignancy and remains the leading cause of cancer-related deaths worldwide. Non-small cell lung cancer (NSCLC) is the most prevalent subtype, accounting for the majority of these fatalities. Despite numerous approved therapies, the five-year survival rate for NSCLC patients remains poor, largely due to drug resistance and early relapse. Three-dimensional (3D) tumor models that better recapitulate the in vivo conditions of primary NSCLC hold significant potential for advancing both drug discovery and development. In this study, we developed a hydrogel-based, 3D-bioprinted NSCLC model in a tumor-slice format, in which tumor spheroids or organoids of NSCLC and primary CAFs were directly assembled to create a model that mimics the structure and biochemical properties of the tumor microenvironment (TME). Within this bioprinted approach the spatial distribution and compartment ratios can be defined, enhancing reproducibility and enabling a customizable TME reconstruction. A composite bioink consisting of alginate, collagen, and Matrigel was developed to generate a bio-functionalized hydrogel matrix that supports the structural stability and enables the co-culture of different cell types. Several drugs used for first-line treatment of NSCLC were applied to the system to evaluate the treatment efficiency. The printed 3D tumor slices can be cultured for up to 14 days while maintaining their architecture and proliferative capacity. The process is highly reproducible and supports consistent generation of tumor-like constructs suitable for downstream analyses and drug testing. Functional assays including cytotoxicity assays, live imaging, multiplex immunofluorescence staining and 3D imaging were established to assess cell viability, tumor-stromal cell interaction, and treatment response within the printed 3D tumor slices. The culture can also be expanded to include PBMCs to investigate immune cell behavior within the TME, making the model suitable for evaluating immunotherapies. In conclusion, we established a standardized 3D hydrogel-printed model incorporating NSCLC tumor spheroids or organoids and primary CAFs. By reconstructing the TME, this system provides a reproducible and physiologically relevant platform for preclinical drug screening and cancer research. Together with the established analytical methods, this platform serves as a valuable tool for drug discovery and development. Citation Format: Jan A. Schlegel, Julia Thiel, Kanstantsin Lashuk, Schueler Julia, Thomas E. Mürdter, Matthias Schwab, Meng Dong. A 3D bioprinted hydrogel-based tumor model of non-small cell lung cancer for preclinical drug testing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 3400.

Efforts to understand microbiome-host interactions in disease development and progression are growing, with short-chain fatty acids (SCFAs) produced by gut bacteria recognized as key metabolic mediators. To facilitate studies on the effect of bacterially secreted metabolites on human cells, we have developed a new LC-MS(/MS) method that allows quantitative analysis of underivatised SCFAs and simultaneous non-targeted screening for additional microbial or endogenous metabolites from cell culture media. Comprehensive method optimization resulted in methanolic gradient elution using a Kinetex XB-C18 column allowing for quantitative QQQ-MS analysis of propionate, butyrate, isobutyrate, valerate, isovalerate, and caproate, using deuterated internal standards. The method was validated in terms of linearity, selectivity, stability as well as inter-and intra-day accuracy and precision. To integrate non-targeted metabolomics, the assay was subsequently transferred to an LC-QTOF-MS platform and cross-validated to ensure accuracy and precision of SCFA analysis. Applying this method to study the effects of the breast cancer drug tamoxifen and 10 human tamoxifen metabolites (desmethyltamoxifen, Z-endoxifen, (E/Z)-4 '-hydroxy-desmethyltamoxifen, (E/Z)-4-hydroxytamoxifen, tamoxifen-Nglucuronide, (E/Z)-tamoxifen-4-glucuronid, (E/Z)-DM-tamoxifen-4-O-glucuronide, (E/Z)-4-hydroxytamoxifen-N-glucuronide, (E/Z)-endoxifen-4-sulfate, (E/Z)-tamoxifen-4-sulfate) on the secretome of faecal bacterial communities revealed a pronounced impact of the parent drug and non-sulfated metabolites. The omission of derivatisation leads to fast and simple sample preparation. Together with the short LC run time (10 min) and the comprehensive information obtained by combining targeted and non-targeted metabolomics in a single run, the new method represents a valuable tool for the in vitro investigation of metabolite-mediated microbiome-host interactions.

Estrogen receptor (ER) positive luminal early breast cancer (BC) benefits from endocrine therapy based on the selective ER modulator tamoxifen (TAM) and aromatase inhibitors (AI). Yet, tumor recurrence occurs in up to one third of patients due to primary and secondary endocrine treatment resistance (ETR). The causes of primary ETR are not well defined by predictive tumor markers, however, TP53 mutations have been linked with primary ETR in a preoperative short-term endocrine treatment study. We used a previously learned TP53 mutation-predictive p53 score as part of the NanoString®BC360 mRNA expression panel to identify patients with p53 mutant-like status, and tested its applicability for the prediction of primary ETR in the adjuvant setting. In the TCGA BRCA cohort (N=1078), the p53 score distribution was analyzed for the presence of different subgroups using Gaussian finite mixture modeling. The same procedure was applied in breast cancer patients from the BreastMark database (N=1936) as an independent test of methodological and clinical validity. An ad-hoc binary classifier was trained to apply the cutoff identified in TCGA patients to patients of a prospectively collected, postmenopausal early breast cancer test cohort (IKP211) of hormone receptor-positive patients treated with TAM and/or AI (median follow up 6 years, N=632; DRKS00000605). Cluster modeling revealed that the p53 score distribution in the TCGA cohort is composed of two different patient groups: 363 patients (33.7%) with higher p53 scores were separated from the remaining patients with lower p53 scores. Likewise, the BreastMark cohort also showed two clusters with 419 patients (21.6%) forming the patient subgroup with high p53 scores. Cancer-specific survival (TCGA) and recurrence-free survival (BreastMark) were significantly shorter in p53-high patients of Luminal A subtype (P=0.0045 and P=0.04, respectively), with no significant differences observed in the remaining subtypes. When applying the binary classifier (mean AUC 0.95) on patients of the IKP211 test cohort, 14 patients (2.2%) were predicted as p53 mutant-like. This p53-positive patient group was strongly associated with higher recurrence or death rates in multivariate Cox regression (HR=3.3, 95% confidence interval 1.5-7.4; P=0.0037). As an independent verification, TP53 amplicon sequencing of these patients confirmed a positive mutation status in all samples that passed quality control (N=11, 100%). We conclude that this p53-based division of Luminal A patients is useful to predict high-risk hormone receptor-positive early breast cancer patients with p53 mutations upfront, whose tumors likely show primary ETR and who are not sufficiently treated with endocrine agents. Florian A. Büttner, Stefan Winter, Johanna Dahlen, Thomas E. Mürdter, Reiner Hoppe, Peter Fritz, Patrick Danaher, Stephen F. Madden, Matthias Schwab, Werner Schroth. Identification of high-risk luminal early breast cancer patients based on a p53 signature [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 728.

The selective estrogen receptor modulator tamoxifen is a mainstay of endocrine breast cancer therapy. However, the clinical response rates of tamoxifen are inferior to those of aromatase inhibitors, which may be partially explained by variable drug exposure due to the pharmacogenetics of the drug-metabolizing enzyme cytochrome P450 (CYP) 2D6. Clinical trials investigating the association between CYP2D6 impairment and tamoxifen outcomes have yielded conflicting results. The results of a comprehensive meta-analysis of 33 single-center tamoxifen trials reported here address this inconsistency by adjusting for two biases that may affect the validity of previous association studies: allele coverage of CYP2D6 genotyping and loss of heterozygosity of the CYP2D6 locus in tumor-derived DNA. After adjustment for bias, meta-analyses show significantly reduced study heterogeneity and a higher risk of recurrence or death in patients with impaired CYP2D6 metabolism compared with those with normal activity. These data may support the use of pharmacogenetics-guided tamoxifen therapy to improve outcomes in patients with CYP2D6-compromised breast cancer. Prospective studies should be considered. See related article by MacLehose et al., p. 224.

PURPOSE::Tamoxifen undergoes bioactivation to its active metabolite (Z)-endoxifen, which blocks estrogen-dependent breast tumor growth at high potency. We tested the feasibility and safety of supplementing standard tamoxifen therapy with low-dose (Z)-endoxifen in patients with breast cancer with compromised tamoxifen bioactivation. PATIENTS AND METHODS::We conducted a prospective, interventional, three group randomized trial including 235 patients with hormone receptor-positive breast cancer who received standard tamoxifen therapy (20 mg/day). Patients were stratified by CYP2D6 genotype (n = 78), defining poor, intermediate, and normal metabolizers, or by baseline (Z)-endoxifen plasma concentration (n = 78), defining ≤15, 15 to 25, and ≥25 nmol/L. Co-treatment with (Z)-endoxifen 3 and 1.5 mg/day or placebo was performed, respectively. A control group (n = 79) received placebo regardless of metabolizer phenotype. The primary endpoint was the number of patients with (Z)-endoxifen levels >32 nmol/L after 6 weeks of treatment. Adverse events were continuously monitored. RESULTS:A higher proportion of patients in both intervention groups achieved target concentrations >32 nmol/L compared with control (P < 0.0001). At 3 mg (Z)-endoxifen supplementation, 92.3% of CYP2D6 poor metabolizer patients and all patients with baseline (Z)-endoxifen ≤15 nmol/L achieved the target concentration. At 1.5 mg (Z)-endoxifen supplementation, 88% of CYP2D6 intermediate metabolizer patients and 95% of patients with 15 to 25 nmol/L baseline (Z)-endoxifen levels achieved the target concentration. Similar proportions of patients receiving (Z)-endoxifen (6/80, 7.5%) or placebo (8/155, 5.2%) experienced grade 3 adverse events. CONCLUSIONS:Adding low-dose (Z)-endoxifen to standard tamoxifen is safe and provides a new approach to personalized antiestrogen treatment for patients with low endoxifen plasma levels.

Clear cell renal cell carcinoma (ccRCC) is characterized by a metabolic shift towards enhanced aerobic glycolysis and increased lactate production. The survival rate for metastatic RCC is still poor. We evaluated the lactate monocarboxylate transporter 4 (MCT4), encoded by SLC16A3, as drug target for metastatic disease. MCT4 protein expression in 209 distant ccRCC metastases, including 40 recurrent metastases, was generally as high as compared to primary tumor tissue and significantly increased compared to non-tumor tissue (P<1E-15). MCT4 expression was irrespective of affected organs and mutations in RCC driver genes. DNA methylation in the SLC16A3 promoter, assessed by MALDI TOF mass spectrometry and correlated with clinicopathological data, were not significantly different in metastases of all investigated organ sites, and between paired tumor and metastases samples. Visualization of expression in single-cell and spatial RNA sequencing datasets reveals main expression of SLC16A3 in cells derived from tumor, tumor-normal interface, metastatic and lymph node tissue. Alone or combined with inhibition of mitochondrial respiration by metformin and phenformin, the MCT4 inhibitor syrosingopine significantly inhibits lactate efflux, induces cell viability reduction in four different RCC cell lines and patient-derived 2D/3D models, and alterations in cellular metabolism and mitochondrial respiration. Six patient-derived RCC air-liquid interface models, mimicking the complex RCC architecture, corroborate these data. Beyond potential prediction of patient outcome using MCT4 expression and DNA methylation at specific CpG sites, drug targeting of MCT4 and inhibiting mitochondrial respiration synergistically is a novel treatment strategy for metastatic ccRCC.

Mantle cell lymphoma (MCL) is a rare, aggressive B-cell neoplasm that frequently relapses and only shows a limited response to conventional chemotherapy. A major challenge in MCL research is culturing primary MCL cells ex vivo, as cells tend to undergo spontaneous apoptosis when cultured in 2D suspension culture. Although 3D models are known to better recapitulate the in vivo situation of solid tumors, their application is still poorly explored in lymphomas. Developing 3D models that replicate the in vivo conditions of MCL within the lymph node could enhance their survival, facilitate the study of the MCL-tumor microenvironment crosstalk, and mimic the in vivo drug response. Here, a 3D printed model of MCL in the form of hydrogel tumor slices was established, along with an optimized culture method. A standardized process was developed using MCL cell lines or primary MCL cells, in which the cells are immersed in a hydrogel containing alginate, type I collagen, and basement membrane matrix by bioprinting into a gelatin support bath. The resulting MCL hydrogel tumor slices are cultured on a filter support to maintain their stability throughout the culture period. Drug treatments can be applied to the system. The response of single cells inside the hydrogel tumor slice can be tracked by four-color live 3D fluorescence imaging. Primary MCL cells demonstrated a stable viability when cultured in the hydrogel tumor slices. This protocol provides a detailed description of the generation, culture, and analysis of MCL cells in hydrogel tumor slices. By closely mimicking the tumor microenvironment and utilizing an air-liquid interface culture, the presented model enhances physiological relevance compared to the traditional 2D culture. It offers significant potential for advancing both biological and therapeutic studies of MCL.

AbstractPurpose :Tamoxifen undergoes Patients and Methods :We conducted a prospective, interventional, three group randomized trial including 235 patients with hormone receptor–positive breast cancer who received standard tamoxifen therapy (20 mg/day). Patients were stratified by CYP2D6 genotype (n = 78), defining poor, intermediate, and normal metabolizers, or by baseline (Z)-endoxifen plasma concentration (n = 78), defining ≤n = 79) received placebo regardless of metabolizer phenotype. The primary endpoint was the number of patients with (Z)-endoxifen levels >32 nmol/L after 6 weeks of treatment. Adverse events were continuously monitored. Results: A higher proportion of patients in both intervention groups achieved target concentrations >32 nmol/L compared with control (P < 0.0001). At 3 mg (Z)-endoxifen supplementation, 92.3% of CYP2D6 poor metabolizer patients and all patients with baseline (Z)-endoxifen ≤15 nmol/L achieved the target concentration. At 1.5 mg (Z)-endoxifen supplementation, 88% of CYP2D6 intermediate metabolizer patients and 95% of patients with 15 to 25 nmol/L baseline (Z)-endoxifen levels achieved the target concentration. Similar proportions of patients receiving (Z)-endoxifen (6/80, 7.5%) or placebo (8/155, 5.2%) experienced grade 3 adverse events. Conclusions: Adding low-dose (Z)-endoxifen to standard tamoxifen is safe and provides a new approach to personalized antiestrogen treatment for patients with low endoxifen plasma levels.

Suppl Fig. S1: Concentration time course of tamoxifen, N-desmentyl tamoxifen, (Z)-4-hydroxy tamoxifen, and (Z)-endoxifen

Gastrointestinal stromal tumours (GISTs) are the most common mesenchymal tumours of the gastrointestinal tract and a key example for targeted therapy with tyrosine kinase inhibitors (TKIs), which have significantly improved survival rates. However, no effective treatments exist for TKI-resistant or mutation-negative tumours. Until now, research on the effects of TKIs has mainly used 2D cultures or mouse models, lacking patient-specific 3D GIST models. We investigated various 3D GIST models, including spheroids, organoids, patient-derived microtumours (PDMs), and precision-cut tumour slices (PCTSs), to assess their feasibility as alternatives for 2D cell culture or in vivo mouse models. Moreover, 2D monolayer and 3D spheroid GIST cell lines showed mutation-dependent responses to TKI treatment, but differences between 2D and 3D cultures were minimal. Thus, patient-derived 3D models, incorporating tumour microenvironment cells, were developed for more accurate in vivo representation. PDMs and PCTSs were successfully isolated from primary tumours and cultivated for up to two weeks. Three-dimensional models were immunohistochemically characterised, and the response to TKI therapies was tested and compared with expected clinical outcomes. In addition to already established 2D cell cultures and mouse models, PDMs and PCTSs are novel patient-derived 3D models that can be used to study tumour cell interactions within the microenvironment. Moreover, they could be used to investigate TKI resistance, and novel treatment options such as immunotherapies and combination therapies.

Gastrointestinal stromal tumors (GIST) are the most prevalent mesenchymal tumor entity of the gastrointestinal tract and serve as a key example for targeted therapies with tyrosine kinase inhibitors (TKI) such as imatinib, in tumors with the most frequent mutations in PDGFRA or c-Kit. Although TKIs have significantly improved the overall survival rates of GIST patients, there are currently no treatment options for imatinib-resistant tumors and tumors without identifiable driver mutations. So far, current sarcoma research mostly relies on expensive mouse models or 2D cell culture that do not integrate the tumor microenvironment (TME). Therefore, clinically relevant, patient-derived 3D GIST models that include cells of the TME are urgently needed. In this study, we tested the feasibility of precision-cut tumor slices (PCTS) as a preclinical model for personalized GIST research. PCTS are patient-derived 3D tissue slices that enable the study of interactions between tumor cells and cells of the surrounding TME. For this, primary GIST tumors were cut into 250µm thick slices and cultured for up to two weeks with or without standard TKI treatment. Tumor morphology and cell composition was analyzed via immunohistochemical (IHC) and multiplex immunofluorescence (mIF) staining, which enabled the simultaneous detection of up to six different markers on one FFPE tissue section. The downstream analysis included a machine-learning based workflow that allowed the comparison of biomarker expression and spatial immune cell distribution in the PCTS stromal and tumor areas before and after treatment. Comparison between primary tumor and PCTS revealed a preserved tumor morphology and heterogeneous cell composition throughout culture. In addition, serum levels of various cytokines were assessed with LEGENDPlex assays and correlated with the PCTS cell composition. The in vitro response of PCTS to TKI-based therapies was analyzed via IHC staining and ATP-based viability assays. PCTS from tumors with PDGFRA mutations were expected to be imatinib-resistant, while PCTS from tumors with c-KIT mutations were considered imatinib-sensitive. Sequencing analyses determined the PDGFRA or c-KIT mutational status showing that all patients harbored either a PDGFRA or a c-KIT mutation. PCTS treatment response was compared to the mutational status and if available to clinical follow-up data. Of all PCTS that responded to imatinib therapy, two cases matched the anticipated clinical outcome. PCTS provide a clinically relevant new model system for GIST to explore the development of therapy response to TKI-based treatment and to investigate novel GIST treatments such as immunotherapies and combination treatments. The work is supported by the Robert Bosch Stiftung, Stuttgart, Germany Dina Mönch, Julia Thiel, Meng Dong, Annika Maaß, Annette M. Staiger, Katrin Kurz, German Ott, Philipp Renner, Tobias Leibold, Thomas Mürdter, Matthias Schwab, Marc-Hendrik Dahlke, Jana Koch. Evaluation of individual drug response to tyrosine kinase inhibitor treatment in precision-cut tumor slices of gastrointestinal stroma tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6845.

Inter- and intra-assay variability

Representativeness of study participants

Colorectal cancer (CRC) constitutes the second leading cause of cancer-related death worldwide and advanced CRCs are resistant to targeted therapies, chemotherapies and immunotherapies. p38α (Mapk14) has been suggested as a therapeutic target in CRC; however, available p38α inhibitors only allow for insufficient target inhibition. Here we describe a unique class of p38α inhibitors with ultralong target residence times (designated ULTR-p38i) that robustly inhibit p38α downstream signaling and induce distinct biological phenotypes. ULTR-p38i monotherapy triggers an uncontrolled mitotic entry by activating Cdc25 and simultaneously blocking Wee1. Consequently, CRC cells undergo mitotic catastrophe, resulting in apoptosis or senescence. ULTR-p38i exhibit high selectivity, good pharmaco-kinetic properties and no measurable toxicity with strong therapeutic effects in patient-derived CRC organoids and syngeneic CRC mouse models. Conceptually, our study suggests ultralong-target-residence-time kinase inhibitors as an alternative to covalent inhibitors, which, because of the lack of cysteine residues, cannot be generated for many kinase cancer targets. Rudalska et al. describe a novel class of p38α inhibitors with increased target residence time. They explore the drugs’ specificity, pharmacokinetics and toxicity profile and show that they are efficacious in the context of colorectal cancer.

Endpoint analysis in PP-population

Suppl Fig S2: Concentration ratio (E)-endoxifen to (Z)-endoxifen

Immunotherapies have emerged as a promising pillar for cancer therapy. However, due to their reliance on the individual, complex and dynamic tumor microenvironment, treatment success remains limited to a subset of patients. A deeper understanding of the response of individual tumors and their heterogeneous microenvironment is essential to improve patient stratification and develop novel treatment approaches. This requires suitable model systems which allow tracking of the tumor microenvironment‘s (TME) response to immunotherapy. In this study, we combined patient-derived precision-cut tumor slices (PCTS) with multiplex immunofluorescence staining and cytokine analyses to trace the treatment response of different cell types in their native TME while preserving their location. PCTS derived from ovarian cancer, lung cancer and colorectal cancer tissue (n=13) were treated with OMTX305, an anti-FAP T-cell engaging bispecific antibody, for up to three days. A co-culture system integrating PCTS with autologous PBMCs was established to mimic the in vivo environment, and enable monitoring of immune cell activities. While treatment responses were patient-specific, determination of overall ATP levels of PCTS showed a significant decrease in viability after three days of treatment. Multiplex immunofluorescence stainings were performed to spatially analyze therapeutic effects, including T cell infiltration and distribution, T cell activation (Granzyme B) and cell death induction (Cleaved Caspase-3) in different cell types. T cells were significantly activated after treatment, and tumor slices showed significant cell death induction on the second day of treatment. Some patients showed T cell infiltration into FAP+ regions in response to treatment. Analysis of over 10 cytokines in the culture medium provides deeper insights into how the treatment induced immune cell activation. By integrating spatial biology and cytokine analysis with our PCTS platform, we can map the relationship between the tumor, stromal, and immune cell components within the TME, visualize inter-patient differences, and predict individual therapy responses to immunotherapy. This makes our platform a valuable tool for personalized cancer medicine. Julia Thiel, Jan Schlegel, Adrian Kneer, Sascha Dreher, Annelie Schäfer, Oliver Seifert, Katrin S. Kurz, Marc-H Dahlke, Georg Sauer, Gerhard Preissler, Laureano Simon, Myriam Fabre, Isabel Egaña, Roland Kontermann, Monilola Olayioye, German Ott, Walter E. Aulitzky, Thomas E. Mürdter, Matthias Schwab, Meng Dong. Advancing personalized medicine using precision-cut tumor slices - assessing individual responses to immunotherapy in the tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1602.

Abstract Ovarian cancer is a complex and heterogeneous disease and the major cause of death among women with gynecological cancers. While patients usually respond well to the platinum-based first-line chemotherapy, the disease often becomes increasingly resistant to the treatment. The tumor microenvironment (TME) plays an important role in tumor development and drug resistance. Characterizing the TME of ovarian cancer after drug treatment is essential to identify molecular mechanisms that allow residual tumor cells to survive chemotherapy in ovarian cancer. We have established a preclinical model using precision-cut tumor slices (PCTS) with 250 µm thickness which preserves the TME of solid tumors with their heterogeneous cell composition during cultivation. The tumor morphology, viability and heterogeneity in terms of tumor and stromal cells as well as the immune compartment are preserved within the system. Based on the PCTS model, we also established a method to perform single-cell RNA sequencing (scRNA-seq) of the PCTS after cultivation and drug treatment to characterize the cellular features and dynamic relationships of different cell populations in the TME after drug treatment. Over 20 cell subtypes including different clusters of epithelial cells, immune cells and fibroblasts can be identified in PCTS after cisplatin treatment through the scRNA-seq analysis. This enables further deeper analysis of each cell subgroup in the TME after drug treatment. In response to cisplatin treatment, patients PCTS showed an individual induction of PD-L1 in different cell types. Additionally, multiplex immunofluorescence (mIF) stainings of the PCTS were performed to spatially trace the response of the TME to drug treatment. The mIF staining allows the simultaneous detection of up to 6 different markers on one FFPE tissue section. Images were analyzed using a machine-learning based workflow, which allows the comparison of biomarker expression and immune cell spatial distribution in the PCTS stromal and tumor areas before and after treatment. Combining the PCTS as a preclinical model with scRNA-seq and mIF staining analysis allows us to bridge the cellular characteristics and cellular spatial distribution with treatment response in the TME of solid tumors. It enables the systematic analysis of residual cancer populations after cisplatin treatment within the complex TME and further identification of predictive markers and targets for an individualized combination of chemotherapy with compounds targeting these mechanisms. The individual drug response of patients can be deeply evaluated and efficacious therapies can be further developed. Citation Format: Julia Thiel, Adrian Kneer, Weimeng Yu, Bernd Winkler, Georg Sauer, German Ott, Walter E. Aulitzky, Thomas E. Mürdter, Matthias Schwab, Chunguang Liang, Meng Dong. Evaluation of individual drug response in tumor microenvironment to cisplatin treatment in precision-cut tumor slices of ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 639.