A patient-derived organoid model identifies TP53-dependent gemcitabine sensitivity in spinal chordoma Authors: Panpan Hu, Shengxin Zeng, Juncai Lei, Hua Zhou, Ben Wang, Zhongjun Liu, Xiaoguang Liu, Feng Wei Published: 2026 ABSTRACT BackgroundChordoma is a rare malignant bone tumor with limited effective systemic treatment options. Conventional chemotherapy generally shows minimal benefit, highlighting the need for predictive preclinical models to explore therapeutic vulnerabilities. Patient-derived organoids (PDOs) have emerged as a promising three-dimensional culture system that preserves tumor architecture and molecular features while enabling functional testing. However, mechanistic studies linking pathway activity to drug response in chordoma PDOs remain limited.MethodsFresh surgical specimens from five patients with primary spinal chordoma were used to attempt generation of three-dimensional PDO cultures, of which primary three-dimensional cultures were successfully established. Organoids were characterized by histology, immunohistochemistry, and quantitative PCR analysis of chordoma-associated markers. Functional drug screening was performed using a panel of clinically relevant agents in a representative PDO model, followed by dose–response testing of gemcitabine. The role of TP53 in drug response was examined using small interfering RNA–mediated knockdown, with assessment of cell viability, Ki-67 expression, and DNA-damage response–related proteins. Statistical analyses were performed using one-way analysis of variance, with p < 0.05 considered statistically significant.ResultsThe established PDOs recapitulated the histopathological and molecular characteristics of their matched primary tumors and maintained stable growth across six passages. Among the screened agents, gemcitabine showed the strongest growth-inhibitory effect in PDO-based functional assays. Dose–response experiments confirmed significant gemcitabine-induced growth suppression. Importantly, TP53 knockdown markedly attenuated gemcitabine-induced cytotoxicity, increased proliferative activity, and reduced activation of DNA-damage response signaling, indicating a TP53-dependent vulnerability.ConclusionsThis study establishes a spinal chordoma PDO platform for functional precision oncology. Our findings identify a TP53-dependent DNA-damage vulnerability engaged by gemcitabine in patient-derived three-dimensional models, supporting biomarker-informed hypothesis generation rather than routine chemotherapy in unselected patients. This PDO-based approach provides a translational framework for exploring pathway-defined therapeutic susceptibilities in rare tumors such as chordoma.Supplementary InformationThe online version contains supplementary material available at 10.1007/s12672-026-05086-x. Background Chordoma is a rare malignant bone tumor arising from remnants of the notochord, accounting for less than 5% of all primary bone malignancies [1–4]. Despite its slow-growing nature, spinal chordoma often presents with locally aggressive behavior and a high rate of recurrence. Surgical resection remains the mainstay of treatment, while effective systemic therapies are limited [5, 6]. Conventional chemotherapy shows minimal benefit, and targeted agents or immune checkpoint inhibitors have yielded only modest clinical responses [7–10]. Current in vitro models—including two-dimensional cell lines—insufficiently preserve the architectural and molecular complexity of primary chordoma tissue [11–14]. In contrast, patient-derived organoids (PDOs) have emerged as a powerful three-dimensional culture system capable of maintaining the morphological, genetic, and phenotypic characteristics of original tumors [15, 16]. PDOs have been successfully applied in colorectal, pancreatic, and breast cancers to model tumor heterogeneity, study treatment resistance, and guide individualized drug selection [17–21]. Recent work has also demonstrated the feasibility of generating chordoma organoids and using them for early-phase drug discovery [22–25]. However, mechanistic studies linking specific signaling pathways to drug response in chordoma PDOs remain limited [22, 26]. Although the genomic landscape of chordoma has been increasingly characterized, recurrent alterations in classical tumor suppressors such as TP53 appear to be uncommon in conventional disease. Large-scale sequencing and whole-genome profiling studies of skull-base and sacral chordomas have consistently highlighted TBXT (encoding brachyury) amplification, CDKN2A/2B loss, SWI/SNF complex gene mutations, and PI3K pathway alterations as the most frequent genomic events, whereas TP53 mutations have not emerged as predominant drivers [27–30]. In contrast, several immunohistochemical studies have reported overexpression of the p53 protein, the gene product of TP53, in a subset of chordomas, and in some cohorts this was associated with higher proliferative activity or adverse outcome, suggesting that dysregulated p53 signaling may contribute to aggressive behavior even in the absence of recurrent TP53 mutations [31–33]. However, these observations remain largely correlative, and direct mechanistic studies linking TP53 pathway status to treatment response in chordoma are still lacking. Prior studies in other solid tumors have shown that TP53 status is a critical determinant of sensitivity to nucleoside analogs such as gemcitabine, which exert cytotoxicity primarily through DNA replication stress and induction of apoptosis [34–36]. Despite this extensive evidence in other malignancies, whether TP53 function modulates gemcitabine sensitivity in chordoma has not been systematically investigated, and no prior work has established a causal relationship between TP53 activity and drug response in chordoma models. To address these gaps, we established a PDO platform derived from freshly resected spinal chordoma specimens to preserve tumor architecture and molecular features while enabling functional testing. Rather than advocating cytotoxic chemotherapy in a generally chemoresistant disease, this platform was designed to uncover unexpected, pathway-defined vulnerabilities and generate testable hypotheses for patient stratification. Using this approach, we evaluated clinically relevant agents and investigated whether TP53 pathway integrity contributes to drug response in chordoma. Methods Patient-derived chordoma organoid culture and passaging Chordoma specimens were processed following previously described protocols [37]. Briefly, chordoma tissues were rinsed three times for 5 min each in ice-cold phosphate-buffered saline (PBS) containing penicillin/streptomycin (Solarbio, P1400), and subsequently minced into 1–3 mm² fragments on ice using sterile instruments. The fragments were subjected to an additional three PBS washes under identical conditions. Enzymatic dissociation was carried out in 10 mL of a digestion cocktail consisting of Advanced DMEM/F12 (Thermo Fisher, 12634028), collagenase IV (500 U/mL; Sigma-Aldrich, C9407), collagenase II (1.5 mg/mL; Solarbio, C8150), hyaluronidase (20 mg/mL; Solarbio, h8030), dispase II (0.1 mg/mL; Sigma-Aldrich, D4693), ROCK inhibitor Y-27,632 (10 µM; Sigma-Aldrich, Y0503), and 1% fetal bovine serum. Tissues were incubated at 37 °C for 30 min on an orbital shaker. The resulting cell suspension was filtered through a 100 μm cell strainer, followed by centrifugation at 200 × g for 5 min. For organoid initiation, approximately 10,000–20,000 viable tumor cells were resuspended in Matrigel and seeded into pre-warmed 24-well flat-bottom culture plates (Costar, 3524). After Matrigel polymerization, 500 µL of freshly prepared chordoma culture medium (Innovation Biotechnology, IB-BCOM-01, Tianjin, China) was added. The medium, based on Advanced DMEM/F12, was supplemented with penicillin/streptomycin, Glutamax, 10 mM HEPES, WNT3a, R-spondin 1, Noggin, Y-27,632, N-acetylcysteine, nicotinamide, recombinant EGF, recombinant FGF10, recombinant HGF, and DAPT. Medium was refreshed every three days, and organoid structures emerged within 2–3 days post-seeding. Regarding the passaging of tumor-derived organoids, in brief, Matrigel-embedded organoids were collected into a 15 mL centrifuge tube using ice-cold PBS and centrifuged at 200 × g. After removal of the supernatant, the pellet was resuspended in ice-cold PBS and mechanically dissociated by pipetting 30–60 times with a 1 mL pipette. This procedure resulted in partial fragmentation into small multicellular aggregates rather than complete dissociation into single cells. The resulting cell suspension was washed and re-embedded in Matrigel at a 1:2 ratio. Passaging was performed at defined intervals based on organoid growth status, and the mechanical dissociation procedure, including pipetting frequency and handling steps, was standardized across experiments to ensure consistency. Culture medium was replaced every three days. For cryopreservation, organoids were frozen in cryoprotective medium (Innovation Biotechnology, IB-FB) and revived using chordoma culture medium supplemented with 10 µM Y-27,632. RNA isolation and reverse transcription quantitative PCR (RT-qPCR) Total RNA from chordoma tissues was extracted from fresh surgical specimens. For organoid experiments, total RNA was extracted from chordoma organoids seeded at an approximate density of 100 organoids per individual well in a 96-well plate. The extraction process was carried out as follows: initially, the chordoma organoids were carefully seeded in a 24-well plate. After a growth period of three days, the organoids were lysed by using TRIzol (purchased from the Beyotime Institute of Biotechnology), strictly adhering to the manufacturer’s provided instructions. Subsequently, cDNA synthesis was executed with the BeyoRT First Strand cDNA Synthesis Kit (with the catalog number D7166, also from the Beyotime Institute of Biotechnology). For the real-time quantitative polymerase chain reaction (qPCR), the BeyoFast SYBR Green qPCR Mix (2X, catalog no. D7260–25 ml, sourced from the Beyotime Institute of Biotechnology) was utilized on a 7500 Fast Real-time PCR System (manufactured by Applied Biosystems, a subsidiary of Thermo Fisher Scientific, Inc.). The qPCR protocol was meticulously designed, commencing with an initial denaturation step at 95 °C for a duration of 2 min. This was followed by 40 cycles, each cycle consisting of denaturation at 95 °C for 15 s, annealing at 60 °C for 1 min, and extension at 72 °C for 1 min. After the completion of the cycling, a final extension was carried out at 72 °C for 10 min. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was selected as the internal reference gene, and the relative gene expression levels were calculated employing the 2-ΔΔCq method as described in reference [38]. The primers employed in this experiment are detailed in (Table 1). To ensure the reliability and reproducibility of the results, this entire experiment was repeated in triplicate. Table 1Primers used in the present studyGenePrimerSequence TBXT SenseACCCAGTTCATAGCGGTGACAnti-senseAGTTGTCAGAATAGGATTGGGAG TP53 SenseAAGTCTAGAGCCACCGTCCAAnti-senseCAGTCTGGCTGCCAATCCA EGFR SenseCCTGAGCTCTCTGAGTGCAAAnti-senseAGGTTGAGGAGCAGGACTGT PDGFRB SenseCCATCAGCAGCAAGGCGAAnti-senseAGCAGGTCAGAACGAAGGTG GAPDH SenseAATGGGCAGCCGTTAGGAAAAnti-senseGCCCAATACGACCAAATCAGAG Table 1Primers used in the present studyGenePrimerSequence TBXT SenseACCCAGTTCATAGCGGTGACAnti-senseAGTTGTCAGAATAGGATTGGGAG TP53 SenseAAGTCTAGAGCCACCGTCCAAnti-senseCAGTCTGGCTGCCAATCCA EGFR SenseCCTGAGCTCTCTGAGTGCAAAnti-senseAGGTTGAGGAGCAGGACTGT PDGFRB SenseCCATCAGCAGCAAGGCGAAnti-senseAGCAGGTCAGAACGAAGGTG GAPDH SenseAATGGGCAGCCGTTAGGAAAAnti-senseGCCCAATACGACCAAATCAGAG Primers used in the present study Hematoxylin and eosin (H&E) staining and immunohistochemistry To evaluate the histological similarity between tumor organoids and the original tumor specimens, tumor samples and chordoma-derived organoids were fixed in 4% paraformaldehyde freshly prepared (Sigma-Aldrich, P6148-1KG), then subjected to standard dehydration and paraffin embedding procedures. Tissue sections of 4.5 μm thickness were obtained, rehydrated, and stained with H&E. For immunohistochemical analysis, endogenous peroxidase activity was inhibited by incubation in 3% hydrogen peroxide for 10 min at room temperature. Antigen retrieval was conducted using EDTA buffer for 20 min in a steam-heated chamber. Subsequently, sections were blocked in 10% donkey serum (Solarbio, SL050) for 1 h and incubated overnight at 4 °C with primary antibodies, including anti-Brachyury (Abcam, ab209665). On the following day, sections were treated with secondary antibodies (GTVision III Detection System/Mo & RB, Gene Tech, GK500710) for 1 h at room temperature. Imaging of both H&E and immunohistochemically stained sections was performed using a Zeiss Axio Imager M2 microscope (Carl Zeiss, Oberkochen, Germany). Lentiviral transduction of chordoma organoids Chordoma organoids were cultured for 5–8 days after embedding in Matrigel to allow initial formation before lentiviral transduction. Organoids were transduced with lentivirus carrying green fluorescent protein (GFP) (Sangon Biotech, Shanghai, China) to assess their suitability for basic research. The lentivirus was used at three different concentrations: 5 × 10^7, 1 × 10^8, and 2 × 10^8 transducing units (TU) per milliliter. Prior to viral exposure, Matrigel was partially disrupted and organoids were gently resuspended to enhance viral accessibility. The organoids were exposed to the lentivirus for 3 days to allow for sufficient transduction. After the transduction period, organoids were washed and cultured in fresh medium, and the organoids were examined under a fluorescence microscope to visualize GFP expression. Fluorescence images were captured using standardized imaging settings. For quantitative analysis of GFP expression, images were processed using ImageJ software to measure the intensity of GFP fluorescence, with at least three biological replicates for each lentivirus concentration to ensure reliability and reproducibility of the results. Drug screening and cell viability assay Chordoma organoids derived from the tissue of case 4 patient were used as a representative model for drug screening experiments (Supplement Table 1). For the drug treatment assay, 96-well plates were pre-coated with 2.5% Matrigel to provide a supportive extracellular matrix. Organoids were counted under a light microscope and manually distributed to obtain approximately 100 organoids per well in 90 µL of culture medium. To minimize variability, organoids within a comparable size range were selected for seeding. After a 24-hour incubation at 37 °C in a humidified incubator with 5% CO₂, the organoids were exposed to drugs at a concentration of 10 µM for 48 h. Subsequently, to evaluate the dose-response of gemcitabine, the organoids were treated with increasing concentrations of gemcitabine (1.25, 2.5, 5, 10, and 20 µM) for an additional 48 h. To evaluate cell viability, 50 µL of CellTiter-Glo Luminescent Cell Viability Assay reagent (Promega Corporation, Australia) was added to each well, followed by a 10-minute incubation to allow for complete cell lysis and luminescence stabilization. Luminescence was measured using a microplate reader, and data were analyzed to generate a dose-response curve. All experiments were performed with three technical replicates (n = 3), and results were presented as mean ± SD. Statistical significance was determined with *p < 0.05, **p < 0.01, and ***p < 0.001. Small interfering RNA (siRNA)–mediated TP53 knockdown The siRNA sequences targeting P53 were designed using siRNA-Target-Finder (GeneScript) and synthesized by Synbio Technologies. The negative control siRNA sequence was 5’-AAGAAGCTCAGGAACGCAGGA-3’, and the P53-targeting siRNA (siP53) sequence was 5’-AAGACTCCAGTGGTAATCTAC-3’. Both siRNAs were transiently transfected into chordoma organoids using FuGENE HD Transfection Reagent (catalog no. E2311; Promega Corporation), according to the manufacturer’s instructions, in a 37 °C cell culture incubator with 5% CO2. Transfection was performed 24 h before the subsequent experiments. Knockdown efficiency was assessed by RT-qPCR and western blot assays, following the study’s protocols. Western blot analysis Organoids were harvested using RIPA buffer (Beyotime, catalog number: P0013C). The total protein in the organoids was extracted and quantified using the Bicinchoninic Acid protein assay (BCA; Beyotime, catalog number: P0009, China). Before western blotting analysis, the protein samples (40 µg each) were electrophoresed on 10% Sodium Dodecyl Sulfate-Polyacrylamide Gels. These gels were transferred to a nitrocellulose membrane (Sigma-Aldrich; Merck KGaA). These membranes were blocked with a solution containing PBS + 5% nonfat milk at room temperature for 1 h. The membranes were then incubated with a primary antibody including TP53 (Abcam, catalog number: ab131442) and actin (Abcam, catalog number: ab6276) at 4 °C, overnight. Subsequently, the membranes were washed with PBST, three times, and then incubated with horseradish peroxidase-conjugated secondary antibody (Thermo Fisher, catalog number: C31460100) for 1 h at RT. The labeled proteins were detected and their concentration was determined by a chemiluminescent imaging system (Tanon 5200 system, Tanon, Shanghai, China). Statistical analysis Data are expressed as mean ± standard deviation (SD). Continuous variables were analyzed using one-way ANOVA with Tukey’s post-hoc test, while categorical variables were assessed using Fisher’s exact test. All statistical analyses were performed with GraphPad Prism 5.0 software (GraphPad Software, Inc.), and a p-value < 0.05 was considered statistically significant. Results Establishment and characterization of patient-derived chordoma organoids Organoid cultures were established from surgical specimens obtained during surgery from therapy-naive chordoma patients, as part of our hospital’s living biobank (Supplement Table 1). The chordoma organoids were cultured according to the protocol illustrated in (Fig. 1A). They were grown in 3D within Matrigel for 18 days, exhibiting the typical spherical morphology with variable diameters (Fig. 1B). Additionally, it was demonstrated that these chordoma organoids could be passaged, maintaining morphology and diameter similar to the primary cultures (Fig. 1C). Under the current protocol, the organoids could be passaged up to six times. Overall, three chordoma organoid lines were successfully cultivated. Fig. 1Culture of chordoma organoids. A Diagram of the process of establishing chordoma organoids from patient tumors. B Bright-field images of chordoma organoids in each culture period. The experiments are repeated in all successfully established PDO lines. Scale bar, 200 μm. C Bright-field images showing the unchanged morphologies of chordoma organoids after being passaged. Scale bar, 200 μm. The experiments are repeated three times Culture of chordoma organoids. A Diagram of the process of establishing chordoma organoids from patient tumors. B Bright-field images of chordoma organoids in each culture period. The experiments are repeated in all successfully established PDO lines. Scale bar, 200 μm. C Bright-field images showing the unchanged morphologies of chordoma organoids after being passaged. Scale bar, 200 μm. The experiments are repeated three times To assess whether chordoma organoids could replicate the molecular characteristics of primary tissue, several marker genes, including TBXT, EGFR, PDGFRB, SMARCB1, PIK3CA, PTEN, and CDKN2A were analyzed using qPCR. The results indicated that the expression levels of these marker genes in chordoma organoids were comparable to those in primary chordoma tissues (Fig. 2A, Supplement Tables 2 and 3). Furthermore, immunohistochemistry confirmed that chordoma organoids recapitulated key tumor-cell histopathological features of the primary chordoma tissue (Fig. 2B); however, stromal components present in the native tumor microenvironment were not retained in the organoid cultures. Additionally, the expression of the chordoma marker protein brachyury was observed in both the organoids and the primary tissues (Fig. 2C). The expression of brachyury appeared to be increased in the chordoma organoids, although no statistically significant difference was observed (Supplementary Fig. S1). Fig. 2Characterization of chordoma organoids. A The expression levels of chordoma tissues and derived organoids marker genes, including TBXT, EGFR, PDGFR, SMARCB1, PIK3CA, PTEN, and CDKN2A (p16). B Hematoxylin and eosin staining images of chordoma original tissues and derived organoids. Scale bar, 100 μm. The experiments are repeated three times. C Immunohistochemistry staining images of chordoma original tissues and derived organoids. The chordoma organoids retained the expression patterns of the characteristic markers (Brachyury). The experiments are repeated three times. Scale bar, 100 μm. *p < 0.05, **p < 0.01, and ***p < 0.001 Characterization of chordoma organoids. A The expression levels of chordoma tissues and derived organoids marker genes, including TBXT, EGFR, PDGFR, SMARCB1, PIK3CA, PTEN, and CDKN2A (p16). B Hematoxylin and eosin staining images of chordoma original tissues and derived organoids. Scale bar, 100 μm. The experiments are repeated three times. C Immunohistochemistry staining images of chordoma original tissues and derived organoids. The chordoma organoids retained the expression patterns of the characteristic markers (Brachyury). The experiments are repeated three times. Scale bar, 100 μm. *p < 0.05, **p < 0.01, and ***p < 0.001 Chordoma organoids support efficient lentiviral transduction To evaluate the feasibility of genetic manipulation in chordoma organoids, we performed lentiviral transduction using GFP-expressing constructs at three viral titers. Organoids exhibited dose-dependent increases in GFP fluorescence following 3 days of exposure (Fig. 3A–B). Fig. 3Lentiviral transduction of chordoma organoids. A Representative image of chordoma organoids transfected with different doses of lentivirus carrying GFP for 3 days. The experiments are repeated three times. Scale bar, 100 μm. B Relative quantification analysis of total GFP fluorescence intensity. *p < 0.05, **p < 0.01, and ***p < 0.001 Lentiviral transduction of chordoma organoids. A Representative image of chordoma organoids transfected with different doses of lentivirus carrying GFP for 3 days. The experiments are repeated three times. Scale bar, 100 μm. B Relative quantification analysis of total GFP fluorescence intensity. *p < 0.05, **p < 0.01, and ***p < 0.001 Chordoma organoids as a platform for drug screening We treated them with a panel of commonly used anti-cancer drugs, including cisplatin, pemetrexed, gefitinib, a combination of cisplatin and paclitaxel, pembrolizumab, afatinib, docetaxel, paclitaxel, and gemcitabine. Our results demonstrated that afatinib, docetaxel, paclitaxel, gemcitabine, and the combination of cisplatin and paclitaxel significantly inhibited the growth of chordoma organoids, with gemcitabine showing the most substantial inhibitory effect among the tested agents (Fig. 4). Fig. 4Drug sensitivity test on the Chordoma organoids. Chordoma organoids were treated with a panel of commonly used anti-cancer drugs, including cisplatin, pemetrexed, gefitinib, a combination of cisplatin and paclitaxel, pembrolizumab, afatinib, docetaxel, paclitaxel, and gemcitabine. n = 3, data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001 Drug sensitivity test on the Chordoma organoids. Chordoma organoids were treated with a panel of commonly used anti-cancer drugs, including cisplatin, pemetrexed, gefitinib, a combination of cisplatin and paclitaxel, pembrolizumab, afatinib, docetaxel, paclitaxel, and gemcitabine. n = 3, data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001 Gemcitabine-mediated growth inhibition is TP53-dependent To further examine gemcitabine sensitivity in chordoma, organoids were treated with increasing concentrations of gemcitabine (1.25–20 µM), resulting in a significant, dose-dependent reduction in organoid viability (Fig. 5A). Fig. 5TP53-dependent gemcitabine sensitivity in chordoma organoids. A Dose-response curve of chordoma organoids to gemcitabine (1.25, 2.5, 5, 10, and 20 µM). n = 3, data are presented as mean ± SD. The expression levels of mRNA (B) and protein (C) in chordoma organoids with TP53 knocked down. n = 3, data are presented as mean ± SD. TP53 knockdown diminished gemcitabine’s inhibitory effects on chordoma organoid growth including expression of proliferation marker, KI67 (D) and cell viability (E). *p < 0.05, **p < 0.01, and ***p < 0.001 TP53-dependent gemcitabine sensitivity in chordoma organoids. A Dose-response curve of chordoma organoids to gemcitabine (1.25, 2.5, 5, 10, and 20 µM). n = 3, data are presented as mean ± SD. The expression levels of mRNA (B) and protein (C) in chordoma organoids with TP53 knocked down. n = 3, data are presented as mean ± SD. TP53 knockdown diminished gemcitabine’s inhibitory effects on chordoma organoid growth including expression of proliferation marker, KI67 (D) and cell viability (E). *p < 0.05, **p < 0.01, and ***p < 0.001 Given the established role of TP53 in mediating DNA-damage responses in other malignancies, we investigated whether TP53 influences gemcitabine sensitivity in chordoma. siRNA-mediated knockdown of TP53 was confirmed at both the mRNA and protein levels (Fig. 5B–C; Supplementary Fig. S2). Notably, TP53 knockdown markedly attenuated the inhibitory effect of gemcitabine, as reflected by increased organoid viability and reduced growth suppression following treatment (Fig. 5D–E). Discussion In this study, we established a panel of patient-derived spinal chordoma organoids that faithfully recapitulated the histological morphology, molecular marker expression, and growth characteristics of their matched primary tumors (Figs. 1 and 2). Beyond demonstrating successful long-term culture, these organoids provided a stable and tractable platform that enabled drug screening, genetic manipulation, and mechanistic interrogation of treatment responses. Using this model, we identified gemcitabine as a top inhibitory hit in PDO-based functional assays (Fig. 4) and uncovered a TP53-dependent basis of gemcitabine sensitivity in chordoma (Fig. 5)—representing, to our knowledge, the first functional evidence linking TP53 activity to chemotherapeutic response in this rare malignancy. Most existing chordoma studies have relied on two-dimensional cell lines or xenografts [11–14], but these models often lose brachyury expression, undergo genetic drift, or fail to reproduce the architecture of native chordoma. Consistent with recent advances in cancer organoid technology [39–42], our PDOs maintained lineage-defining markers such as TBXT (encoding brachyury), EGFR, PDGFRB, PTEN, and CDKN2A, and preserved tumor histoarchitecture over multiple passages (Fig. 2). Compared with conventional models, these PDOs thus provide a more physiologically relevant system for studying chordoma biology and for evaluating patient-specific therapeutic vulnerabilities. Recent work by Al Shihabi et al. demonstrated that chordoma PDOs are feasible for early-phase drug discovery [22], but mechanistic studies have remained limited. Our establishment of three independent spinal chordoma PDO lines, along with the demonstration of robust transduction and gene knockdown, expands the available repertoire of experimental tools and supports the broader use of PDOs for functional genomics and pathway analysis in chordoma. In addition to drug screening, we validated the potential of chordoma organoids as a promising platform for studying lentiviral transfection and the mechanisms of drug treatment responses. Our study successfully transduced chordoma organoids with lentivirus carrying GFP, achieving efficient expression that scaled with viral dose (Fig. 3). This finding is consistent with prior studies in other organoid systems, where lentiviral vectors have been shown to stably integrate and express transgenes [42]. This capability allows for the introduction of specific genetic modifications, facilitating functional studies and drug testing that require gene expression manipulation. Chordoma remains a chemoresistant tumor, and conventional agents have shown limited clinical benefit [5, 6]. In this study, we do not propose gemcitabine as a routine treatment for unselected patients. Instead, PDO-based functional testing offers a means to identify unexpected vulnerabilities in rare tumors with limited systemic options. In our drug panel, several agents—including afatinib, paclitaxel, and docetaxel—exhibited variable cytotoxicity, but gemcitabine produced the most consistent and pronounced inhibition in our representative PDO models (Fig. 4). The emergence of gemcitabine as a top hit in our model suggests that a subset of chordomas may harbor an exploitable DNA-damage response dependency. This observation complements prior work demonstrating the utility of PDO-based drug profiling in predicting treatment response in colorectal, pancreatic, and other cancers [17–21, 39–41, 43]. Gemcitabine has not been systematically evaluated in chordoma, and current clinical experience indicates that conventional cytotoxic chemotherapy generally provides limited benefit in this disease [5, 6]. In a retrospective Japanese study of gemcitabine plus docetaxel in advanced bone and soft-tissue sarcomas, only one patient had chordoma and achieved stable disease [44], but the mixed-histology design and absence of subtype-specific analysis preclude any conclusions regarding chordoma responsiveness. Overall, dedicated clinical evidence for gemcitabine in chordoma remains lacking. In this context, our findings provide preclinical support for gemcitabine as a candidate worthy of further investigation, particularly in biomarker-selected or mechanistically informed settings. Although TP53 mutations are relatively uncommon in conventional chordoma [27–30], several studies have reported aberrant p53 protein accumulation or pathway dysregulation in subsets of tumors, sometimes associated with higher proliferative activity or worse clinical outcomes [31–33]. Importantly, attenuation of cytotoxicity following TP53 knockdown indicates that TP53 pathway integrity, rather than TP53 mutation frequency, is a critical determinant of gemcitabine response. This functional distinction underscores the value of organoid-based assays, which assess pathway activity directly and may reveal therapeutic susceptibilities not predicted by genomics alone. In other malignancies, TP53 is a well-recognized determinant of gemcitabine response, mediating apoptosis, replication stress, and DNA-damage signaling [34–36]. Our work demonstrates that TP53 knockdown in chordoma organoids markedly reduces gemcitabine-induced growth suppression, dampens downstream DNA-damage responses, and preserves proliferative activity (Fig. 5). These results support a model in which gemcitabine sensitivity in chordoma is at least partly dependent on intact TP53 signaling, and they support a testable hypothesis that TP53 pathway integrity may enrich for gemcitabine-responsive chordoma models, warranting further validation. It is important to note, however, that gemcitabine exerts its cytotoxic effects predominantly during the S phase of the cell cycle. Given that conventional chordomas are typically characterized by a low mitotic index in vivo, the sensitivity observed in PDO cultures was somewhat unexpected and warrants careful interpretation. One possibility is that in vitro culture conditions may alter proliferative dynamics compared with the original tumor tissue. Comparative evaluation of proliferation markers (e.g., Ki67) between primary surgical specimens and matched PDOs would provide further mechanistic insight into this question and represents an important direction for future investigation. Despite the promising results obtained from the use of chordoma organoids in this study, several limitations remain that need to be addressed. First, although our PDOs preserve key tumor features, they do not fully incorporate the stromal, immune, and vascular components of the chordoma microenvironment. Future co-culture systems or microfluidic organoid-on-chip models may provide more comprehensive insights. Second, while TP53 knockdown offers strong evidence for a mechanistic role in gemcitabine sensitivity, several factors limit the broader generalizability of this finding. All organoids in this study were derived from primary cervical spine chordomas, and validation was performed in a limited number of PDO lines. Therefore, additional upstream and downstream pathway analyses, as well as validation in sacral, other mobile spine, and recurrent specimens, will be necessary to determine the robustness and broader translational applicability of these observations. Third, prospective correlation of PDO drug sensitivity with actual patient treatment outcomes was not feasible in this study but represents a crucial step toward clinical translation. Finally, large-scale validation of culture media components and long-term genomic stability will be essential for broad implementation of this model. Conclusions Our study establishes a robust spinal chordoma PDO platform that enables drug testing, genetic manipulation, and mechanistic investigation. We identify a TP53-dependent DNA-damage vulnerability engaged by gemcitabine in patient-derived 3D models of spinal chordoma. The present study provides a mechanistically informed, testable hypothesis for biomarker-guided exploration. Future work should integrate multi-omics profiling, microenvironment modeling, and clinical correlation studies to refine this biomarker-guided treatment strategy and improve therapeutic outcomes for patients with chordoma. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3.