Tumor organoids are self-organizing, three-dimensional cultures derived directly from patient tissue, typically embedded in an extracellular matrix scaffold such as Matrigel or synthetic hydrogels and nourished with carefully tuned cocktails of growth factors, including Wnt, EGF, and FGF ligands, alongside inhibitors that steer signaling pathways toward epithelial growth. Unlike conventional two-dimensional cell lines, which accumulate mutations and lose tissue identity over repeated passages, organoids preserve the genetic heterogeneity of the original tumor and reproduce patient-specific phenotypes with remarkable fidelity. This property underpins their most celebrated application: functional drug sensitivity testing, in which a patient’s own organoids are exposed to candidate therapies before clinicians commit to a treatment regimen. Studies across colorectal cancer, pancreatic ductal adenocarcinoma, non-small cell lung cancer, hepatocellular carcinoma, and other tumor types have demonstrated that organoid drug responses can mirror clinical outcomes, offering a living avatar of the disease that static genomic profiling cannot provide.

Yet the review is candid about the limits of conventional organoid culture. Most notably, standard protocols strip away the tumor microenvironment, the intricate ecosystem of cancer-associated fibroblasts, immune cells, endothelial cells, and extracellular matrix that surrounds a tumor in the body. Without this ecosystem, organoids cannot faithfully model immune checkpoint inhibitor responses, antibody-dependent cellular cytotoxicity, or the metabolic crosstalk that fuels resistance. They also flatten spatial heterogeneity, the patchwork of distinct tumor subclones and stromal niches that coexist within a single lesion, and they lack the dynamic physiological forces, such as fluid shear stress, oxygen gradients, and peristalsis-like mechanical cues, that shape tumor behavior in vivo. Add to this a persistent lack of clinical standardization, with laboratories using divergent media formulations, scaffolds, and readouts, and the result is a platform with enormous promise but uneven reproducibility.

The authors’ framework responds by sorting the technological arsenal into four complementary layers. The first is a genetic engineering layer, built on CRISPR-based gene editing and related tools that allow researchers to introduce or correct specific mutations within organoids. This capability transforms organoids from passive surrogates into experimentally tractable systems in which individual genetic alterations can be tested for causality, driver mutations can be validated, and isogenic controls can be generated to isolate the effect of a single variant. Gene-edited organoids have become indispensable for dissecting how specific mutations confer drug resistance, and they open the door to corrective strategies that could one day inform therapeutic design.

The second layer addresses the cellular ecosystem. Co-culture techniques now allow organoids to be grown alongside cancer-associated fibroblasts, endothelial cells such as human umbilical vein endothelial cells, mesenchymal stem cells, and immune populations including peripheral blood mononuclear cells, tumor-infiltrating lymphocytes, natural killer cells, and chimeric antigen receptor T cells. Air-liquid interface culture methods have proven particularly valuable, enabling long-term preservation of immune components within tumor organoids and permitting the modeling of immunotherapy responses that conventional submerged cultures cannot capture. By reconstructing these cellular partnerships, researchers can study immune checkpoint inhibitor activity, antibody-dependent cellular cytotoxicity, and immunogenic cell death in a controlled, patient-specific setting, a capability the review identifies as central to extending organoids into immuno-oncology.

The third layer is spatial and architectural. Single-cell RNA sequencing and spatial multi-omics technologies allow researchers to map the identity and location of every cell type within an organoid, resolving the subclonal architecture and stromal niches that bulk measurements average away. Meanwhile, 3D bioprinting introduces precise control over tissue architecture, enabling the deposition of cells and matrix in defined patterns that recreate vascular structures, stromal compartments, and tumor-stroma interfaces with spatial fidelity. Together, these approaches restore the dimension that traditional organoid culture sacrifices most: the organized, heterogeneous geography of a real tumor, where position within the lesion often determines a cell’s metabolic state, proliferative capacity, and drug sensitivity.

The fourth and most technologically ambitious layer is dynamic systems, embodied by organ-on-a-chip platforms. These microfluidic devices perfuse culture media through channels surrounding organoid tissues, reproducing physiological fluid flow, interstitial pressure, and controlled delivery of nutrients, drugs, and immune cells. Perfusion establishes gradients of oxygen and metabolites that mimic the hypoxic cores and well-vascularized rims of real tumors, while also enabling pharmacokinetic and pharmacodynamic modeling, including absorption, distribution, metabolism, and excretion considerations that static cultures cannot address. By coupling organoids to chips that simulate adjacent organs, researchers can begin to model systemic drug behavior, toxicity, and multi-tissue responses within a single experimental apparatus.

Having assembled the framework, the review maps its applications across four domains. In mechanistic cancer research, integrated organoid platforms allow the dissection of epithelial-mesenchymal transition, metabolic reprogramming between oxidative phosphorylation and glycolysis, and evolutionary trajectories of drug resistance under conditions that approximate the tumor’s native context. In drug discovery, organoid panels serve as high-content screening platforms that filter candidate compounds through patient-relevant biology before animal testing, improving translational success rates. In immunotherapy modeling, immune-competent co-cultures enable preclinical evaluation of checkpoint inhibitors, CAR-T cell therapies, and bispecific antibodies against a patient’s own tumor. In precision clinical decision support, patient-derived organoids function as drug-testing avatars, with emerging evidence linking organoid-predicted responses to pathological complete response in neoadjuvant settings across colorectal, pancreatic, breast, and other cancers.

The authors do not shy away from the obstacles standing between these capabilities and routine clinical use. Clinical translation of organoid-guided therapy remains constrained by the time required to establish cultures, the cost and labor of personalized protocols, variable engraftment and success rates across tumor types, and the absence of standardized operating procedures and prospective, randomized validation trials. The review emphasizes that most published studies pair organoids with one complementary technology at a time, and that the field still lacks a shared logic for combining layers. Its functional integration framework is offered precisely as that logic: a way to decide which capabilities a given biological or clinical question demands, and which technologies should be layered together to meet them, rather than adopting each new tool in isolation.

Looking forward, the review introduces a concept it calls the functional digital twin, a vision in which a patient’s tumor is represented not only by its physical organoid counterpart but by a computationally integrated model that fuses multi-omic data, functional drug responses, and dynamic physiological simulation. Such a twin would allow clinicians to rehearse treatment strategies computationally, predict resistance before it emerges, and refine therapy iteratively as the tumor evolves. Realizing that vision will require advances in standardization, automation, artificial intelligence-driven data integration, and ethical frameworks governing the use of patient-derived models. But the trajectory is clear, the authors argue: as the genetic, ecosystem, spatial, and dynamic layers converge, the humble organoid is evolving from a simplified tumor mimic into a comprehensive experimental platform capable of carrying precision oncology from retrospective correlation to genuinely predictive, patient-specific medicine.

Subject of Research: Tumor organoid technologies and their integration for precision oncology

Article Title: Next-generation tumor organoids: a functional integration framework for advancing precision oncology

Article References: Next-generation tumor organoids: a functional integration framework for advancing precision oncology. (n.d.). https://doi.org/10.1186/s12967-026-08698-7

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08698-7

Keywords: tumor organoids, precision oncology, tumor microenvironment, gene editing, organ-on-a-chip, 3D bioprinting, spatial multi-omics, immunotherapy modeling, patient-derived organoids, drug sensitivity testing, functional digital twin, clinical translation

The raw material for the new models is malignant ascites, the fluid that accumulates in the peritoneal cavity when cancer cells seed the abdominal lining. Ascites is clinically accessible through routine paracentesis, and it is rich in viable tumor cells shed from peritoneal deposits, making it an attractive biopsy surrogate for patients who are too advanced for surgery. Conventional two-dimensional cell cultures, however, strip away the architecture and cell-cell interactions that define how tumors behave, and patient-derived xenografts take months to establish in mice, far too slow to inform a first-line treatment decision. The researchers set out to bridge that gap with extrusion-based three-dimensional bioprinting, a technique that deposits living cells suspended in a soft hydrogel in precise, layered patterns.

The bioink at the heart of the system combines two photo-crosslinkable materials: gelatin methacryloyl, known as GelMA, and hyaluronic acid methacryloyl, or HAMA. GelMA is derived from gelatin and carries methacrylate groups that polymerize under light, giving the printed construct mechanical stability while its cell-adhesive motifs, inherited from native collagen, allow tumor cells to attach, migrate and organize. HAMA contributes a hydrated, hyaluronic-acid-rich matrix that mimics components of the tumor microenvironment and helps regulate stiffness and porosity. A photoinitiator called lithium phenyl-2,4,6-trimethylbenzoylphosphinate triggers crosslinking under visible light during printing, so the structure solidifies gently enough to keep cells alive. Ascites-derived cells were mixed directly into this hydrogel blend, and the resulting ink was printed into small constructs in a process the authors report took less than six hours from sample processing to finished model.

Speed matters because the clinical window for treatment decisions in metastatic gastric cancer is narrow. The printed constructs proved remarkably self-organizing: within seven to ten days in culture, the embedded malignant cells reconstituted organized glandular-like structures or solid tumor architectures, echoing the histological patterns of the parent tumors. The models remained viable for at least two weeks, a sufficient window to run full dose-response experiments. In 13 of 14 patient cases, or 92.9 percent, the team successfully established a bioprinted model, a success rate that compares favorably with organoid cultures and far exceeds the yield of xenograft approaches for this disease stage.

Establishing that the models faithfully represent each patient’s cancer was the central analytical challenge. The researchers performed immunohistochemistry on the printed constructs and found preservation of core gastric cancer markers, including the intestinal lineage transcription factor CDX2 and the carcinoembryonic antigen CEA. More therapeutically important, the expression patterns of three clinically actionable biomarkers, HER2, PD-L1 and Claudin 18.2, were generally consistent between the bioprinted models and the parental tumors. These are the targets that determine eligibility for HER2-directed antibodies, immune checkpoint inhibitors and Claudin 18.2-directed therapies, so their retention in the model is a prerequisite for any claim that the platform can guide treatment selection.

Genomic fidelity was assessed with whole-exome sequencing. The team compared somatic single-nucleotide variants detected in the bioprinted models with those in the parental ascites samples and found that key genomic features were preserved. To quantify this, they correlated the variant allele frequencies of mutations shared between each model and its parent sample, obtaining a median Pearson correlation coefficient of 0.817, with values ranging from 0.679 to 0.921. Variant allele frequency reflects the proportion of sequencing reads carrying a mutation and serves as a proxy for the cellular composition of a sample, so a high correlation indicates that the printed construct recapitulates not just which mutations are present but in what proportions. For a model built from a heterogeneous fluid containing tumor cells, immune cells and stromal elements, that degree of concordance is a strong indicator of faithful representation.

With fidelity established, the platform was put to its intended use: drug sensitivity testing. The researchers exposed each patient’s model to the standard chemotherapeutic backbone for gastric cancer, including oxaliplatin, fluorouracil and paclitaxel, and measured viability across a range of concentrations. Dose-response curves were fitted with a four-parameter logistic model, and the area under the curve, or AUC, was calculated as a summary metric of drug response, with lower values indicating greater sensitivity. The results revealed marked inter-patient heterogeneity: models from different patients responded very differently to the same agents, mirroring the unpredictable treatment responses seen in the clinic and underscoring why one-size-fits-all regimens so often fail in peritoneal disease.

To translate continuous dose-response data into a clinically interpretable verdict, the team needed a decision boundary. They derived an exploratory normalized AUC threshold of 0.675 using the Jenks natural breaks method, a statistical classification technique that identifies natural groupings within a data distribution by minimizing within-class variance and maximizing between-class variance. Applying this threshold, each model’s response to each drug was classified as sensitive or resistant. When the researchers compared these in vitro classifications with the clinical outcomes of the corresponding patients, the results were generally consistent, suggesting that the printed models captured enough of each tumor’s drug biology to predict, at least in an exploratory sense, whether a given agent would help or fail.

The authors are careful to frame the threshold as exploratory rather than validated. A single-arm study of 14 patients cannot establish predictive accuracy with the statistical rigor required for clinical deployment, and the concordance between in vitro classification and clinical outcome will need to be confirmed in larger, prospective cohorts with predefined endpoints such as progression-free survival and overall survival. Questions also remain about how the ascites microenvironment, including immune and stromal cells that may or may not survive the printing process, influences drug responses, and whether the platform can be extended to targeted agents and immunotherapies beyond cytotoxic chemotherapy. The open-access study was conducted under approval from the Medical Ethics Committee of Peking Union Medical College Hospital with written informed consent from all participants, and the authors declare no competing interests.

Even with those caveats, the work represents a meaningful advance in translational cancer modeling. It demonstrates that a clinically obtainable fluid, processed within hours into a bioprinted three-dimensional construct, can reproduce the histological, biomarker and genomic identity of a patient’s gastric cancer and yield drug-response data on a timescale compatible with real treatment decisions. If validated at scale, ascites-derived bioprinted models could give oncologists caring for patients with peritoneal metastasis something they currently lack: a personalized, functional readout of which chemotherapy is likely to work before the first dose is given. The study was supported by the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences, the National High Level Hospital Clinical Research Funding, the Peking Union Medical College Hospital Talent Cultivation Program and the Ningbo Major Research and Development Plan Project.

Subject of Research: 3D bioprinted patient-derived ascites models for drug sensitivity testing in gastric cancer with peritoneal metastasis

Article Title: 3D bioprinted ascites-derived models for gastric cancer patients with peritoneal metastasis

Article References: Hua, Y., Du, L., Sun, H., Sun, M., Pang, M., Jiang, S., Zhang, K., Lu, Y., Mao, Y., Sun, Z., Ge, Y., Nie, M., Wang, C., Wang, X., Yao, G., Yang, D., Bai, C., Yang, H., & Zhao, L. (2026). 3D bioprinted ascites-derived models for gastric cancer patients with peritoneal metastasis. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08953-x

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08953-x

Keywords: 3D bioprinting, gastric cancer, peritoneal metastasis, malignant ascites, drug sensitivity testing, bioink, GelMA, HAMA, whole-exome sequencing, patient-derived models, precision oncology, hydrogels

The clinical stakes are considerable. Although meibomian gland carcinoma accounts for only about 5 to 10 percent of eyelid tumors, reported mortality rates range from 15 to 25 percent, a striking figure for so uncommon a disease. The tumor arises from the meibomian glands, modified sebaceous structures embedded in the tarsal plate of the eyelid, and it behaves badly in every sense. It grows in a multifocal pattern, spreads along the conjunctival epithelium in a characteristic skip or jump fashion that leads surgeons to underestimate its true extent, and frequently returns after resection with positive margins. Once it reaches lymph nodes or distant organs, five-year disease-specific survival falls below 30 percent. Clinicians widely regard it as the most difficult eyelid tumor to diagnose and manage.

Part of the difficulty is that the tumor’s lineage identity is fragile outside the body. Meibomian gland cells carry lipid-rich cytoplasm, terminal sebaceous differentiation programs, and tightly regulated androgen receptor-dependent transcription networks. Under conventional two-dimensional culture conditions these features collapse quickly, producing dedifferentiated cells that no longer resemble the tumor the surgeon removed. The only published human meibomian gland carcinoma cell line, established by Zhang and colleagues, could be maintained for roughly 20 passages before fibroblast overgrowth took over, a familiar failure mode of primary culture. Most of what is known about the disease therefore comes from retrospective clinicopathological studies and static immunohistochemistry of formalin-fixed specimens, which capture structure but not function.

Therapeutic progress has been equally constrained. Conventional chemotherapy agents, including platinum compounds, taxanes, 5-fluorouracil, and mitomycin C, achieve modest response rates and rarely deliver durable control in advanced disease. Genomic profiling studies have identified recurrent alterations in PTGS2, HER2, PIK3CA, and PTEN, along with Hedgehog signaling changes and DNA mismatch repair defects, but without functional platforms to validate these findings, they have not translated into targeted therapies. The authors argue that the absence of scalable, patient-derived models preserving lineage identity, tumor heterogeneity, and functional plasticity has become the central bottleneck in the field, a greater obstacle than the lack of genomic data itself.

Conditional reprogramming, first described by Liu and colleagues in 2012, offers a way around this bottleneck. The technique co-cultures primary epithelial cells with gamma-irradiated mouse 3T3-J2 feeder fibroblasts in F/2 medium supplemented with the Rho-associated kinase inhibitor Y-27632. ROCK inhibition prevents anoikis and cytoskeletal tension-induced apoptosis, while the feeder layer supplies paracrine signals and extracellular matrix components that sustain epithelial survival. Together these inputs converge on pathways governing actin dynamics, cell-cell adhesion, and stress responses, effectively decoupling proliferative capacity from terminal differentiation. The result is a reversible, non-neoplastic, stem-like proliferative state: cells divide furiously, yet they retain lineage-specific transcriptional programs and genomic integrity, and they are not driven toward malignant transformation.

The numbers are remarkable. Conditional reprogramming cultures expand 10 to 100 times faster than conventional epithelial cultures, yielding between 100 million and a billion cells from a 2-millimeter biopsy in under three weeks. Multiple studies have documented high genomic fidelity across extended passages, including preserved variant allele frequencies and an absence of de novo chromosomal aberrations. Equally important is reversibility: remove the feeder layer or the ROCK inhibitor and the cells promptly exit the proliferative state, re-engage native differentiation programs, and rebuild tissue-specific architecture. This distinguishes the method from induced pluripotent stem cell reprogramming and oncogene-based immortalization, both of which erase lineage memory. The system can also generate matched tumor and adjacent normal epithelial cultures from the same patient, creating isogenic pairs that minimize inter-individual genetic variability and allow direct comparison of oncogenic signaling and drug responses.

Over the past decade the platform has proven itself across epithelial cancers. In bladder cancer, Kettunen and colleagues showed that ex vivo drug-response profiles from conditionally reprogrammed cells concorded with clinical outcomes in a personalized screening platform paired with patient-derived xenografts. In prostate cancer, paired tumor and normal cultures revealed tumor-selective synergy between docetaxel and radiation, defining a therapeutic window that spares healthy epithelium. In colorectal cancer, a dose-response matrix approach assessed combined sensitivity to 5-fluorouracil and oxaliplatin, marking the first functional assay to guide adjuvant chemotherapy selection in that disease. Large-scale biobanks built with the technology, including more than 120 head and neck squamous carcinoma lines coupled to whole-exome sequencing and a gastric cancer repository spanning 32 patients, demonstrate its scalability for rare specimens.

Ophthalmic applications have been slower to arrive but are encouraging. Conditionally reprogrammed human limbal epithelial cells have maintained normal karyotype and stable drug-toxicity profiles across passages, validating their use in ocular surface safety screening. The technology has converted human fibroblasts into retinal pigment epithelium-like cells within two weeks, far faster than pluripotent stem cell protocols, and has sustained papillomavirus-positive canine eyelid epithelium for antiviral testing. In conjunctival fornix epithelial cells, the mucolytic drug bromhexine stimulated MUC5AC secretion and lipid-droplet production, hinting at therapeutic targets for ocular lubrication disorders. Most relevant to eyelid oncology, Lee and colleagues used conditional reprogramming to generate the first authenticated trio of ocular adnexal sebaceous carcinoma lines, preserving patient-specific mutations, adipophilin-positive lipid vacuoles, and MYC overexpression, with full validation by short tandem repeat profiling and next-generation sequencing within five passages. Functional testing of mitomycin C, 5-fluorouracil, and the glutamine-metabolism inhibitor DON each induced dose-dependent apoptosis and differentiation, an immediate proof of concept in a historically model-poor disease.

For meibomian gland carcinoma specifically, the review describes a practical workflow. Fresh surgical specimens are minced and enzymatically dissociated, then seeded onto irradiated 3T3-J2 feeder layers in Y-27632-supplemented medium. Validation relies on immunophenotypic markers including cytokeratins, epithelial membrane antigen, androgen receptor, adipophilin, Ki-67, and p63, alongside functional assays in which withdrawal of the ROCK inhibitor tests whether lipid metabolism and sebaceous differentiation can be restored. The authors also stress structured quality control: short tandem repeat authentication, histological correlation with the parental tumor, confirmation of driver alterations and copy-number profiles, longitudinal tracking of tumor-specific variants, and defined passage limits. Such measures guard against the field’s known pitfalls, including preferential expansion of non-malignant epithelium, clonal selection in small biopsies, and possible epigenetic drift during prolonged culture.

Limitations remain real. The mouse feeder layer introduces interspecies signaling that complicates interpretation, and the cultures lack the immune, stromal, and vascular components of the native tumor microenvironment, restricting studies of immune evasion and therapy-induced immune modulation. The authors propose mitigation through pathology-guided sampling, tumor-cell enrichment, molecular monitoring, and the development of feeder-free and hybrid co-culture systems. Looking ahead, they envision conditional reprogramming serving as the upstream expansion step that feeds organoid culture, patient-derived xenografts, multi-omics profiling, and artificial intelligence-assisted analysis of drug-response matrices and imaging phenotypes. In a disease too rare for randomized trials, where surgical specimens are small and fragmented and engraftment rates in xenograft models are low, that combination could anchor functional precision oncology, guide eye-sparing treatment strategies, and finally give clinicians a way to test therapies on living cells from the very tumors they are trying to beat.

Subject of Research: Use of conditional reprogramming cell culture technology to develop patient-derived preclinical models of meibomian gland carcinoma and other eyelid tumors.

Article Title: Conditional reprogramming in eyelid tumors: bridging the model gap in meibomian gland carcinoma

Article References: Conditional reprogramming in eyelid tumors: bridging the model gap in meibomian gland carcinoma. (n.d.). https://doi.org/10.1007/s44272-026-00057-3

Image Credits: AI Generated

DOI: 10.1007/s44272-026-00057-3

Keywords: conditional reprogramming, meibomian gland carcinoma, eyelid tumors, sebaceous carcinoma, patient-derived models, ROCK inhibitor, precision oncology, drug sensitivity testing, ocular oncology, organoids, tumor biobanking, translational research

Patient-derived organoids, or PDOs, are three-dimensional cultures grown directly from a patient’s own tumor tissue, often obtained through surgery, biopsy, or fine-needle aspiration. Embedded in basement membrane extract and nourished with tailored media, dissociated tumor cells reorganize themselves into miniature structures that recapitulate key genetic, histological, and functional features of the original cancer. Unlike conventional two-dimensional cell lines, which accumulate adaptations over years of laboratory passage, PDOs retain the copy-number variations, mutations, and gene-expression programs of the tumor from which they came. Unlike patient-derived xenografts implanted in mice, they can be established and drug-tested within weeks, at a scale that permits longitudinal study of the very same tumor over many treatment cycles. That combination of fidelity and throughput is precisely what makes them attractive for studying the dynamics of treatment failure.

The review’s most distinctive contribution is a dynamic conceptual map that integrates four determinants of drug response — genetic background, cell-state plasticity, cell-cycle state, and microenvironmental context — along three experimentally measurable dimensions: drug sensitivity, proliferative activity, and temporal reversibility. The authors emphasize that clonal evolution, drug-tolerant persister states, cancer stem cell plasticity, cell-cycle heterogeneity, and microenvironment-mediated protection have traditionally been discussed as separate mechanisms of resistance. In reality, they intersect and reinforce one another. A tumor cell may tolerate a targeted drug by entering a reversible, slow-cycling persister state; over months, those persister populations can provide the raw material for stable, genetically fixed resistance. Organoids, because they allow repeated drug exposure, clonal tracking, and single-cell profiling over time, make it possible to watch that transition happen rather than merely infer it from end-stage samples.

The distinction between tolerance, persistence, and frank resistance carries real clinical weight. Tolerance describes a reversible, non-genetic state in which cells survive drug exposure without dividing — the hallmark of drug-tolerant persisters, a subpopulation first characterized in melanoma and lung cancer models treated with targeted agents. Persistence describes the survival of such cells over longer periods and their eventual re-expansion. Resistance, by contrast, typically involves durable changes, such as secondary mutations in drug targets, activation of bypass signaling pathways, or selection of pre-existing resistant clones. The authors caution against conflating these categories, because interventions that work for one may fail for another. A reversible persister state might be dismantled by epigenetic inhibitors given alongside the primary therapy, whereas a genetically resistant clone demands a fundamentally different strategy. Organoids, which preserve both the genetic and the epigenetic heterogeneity of the source tumor, are uniquely suited to distinguishing the two in an individual patient’s cancer.

Technically, the toolkit available to organoid researchers has matured dramatically. Longitudinal drug exposure protocols now allow researchers to mimic the pulsed, cyclical nature of clinical chemotherapy regimens rather than applying constant drug concentrations. Clonal tracking with fluorescent barcoding reveals which lineages survive treatment and repopulate the culture. Single-cell RNA sequencing and single-cell ATAC sequencing resolve the transcriptional and chromatin states of surviving cells, exposing the epithelial-mesenchymal plasticity and stem-like programs that often accompany drug tolerance. Fluorescence ubiquitination cell cycle indicator, or FUCCI, imaging makes cell-cycle heterogeneity visible in real time, linking quiescent subpopulations to survival under cytotoxic pressure. Growth-rate-corrected pharmacological metrics, such as the GR50 and growth-rate-corrected area under the curve, correct for the confounding effects of differing proliferation rates, yielding more comparable drug sensitivity measurements across cultures. Together, these methods turn a static drug-screening assay into a dynamic record of evolutionary response.

Crucially, the review resists the temptation to treat chemotherapy, targeted therapy, and immunotherapy as interchangeable contexts. Evidence from cytotoxic chemotherapy, the authors note, reflects selection among genotypically diverse clones and the protection of quiescent cells, whereas targeted therapy failures frequently involve adaptive rewiring of signaling networks such as EGFR or downstream bypass pathways in a defined genetic background. Immunotherapy and cellular therapy, including CAR-T and CAR-NK approaches, introduce an entirely different dimension: the interplay between malignant cells and immune effectors. Organoid systems are only beginning to capture that dimension through co-culture with tumor-infiltrating lymphocytes, peripheral blood mononuclear cells, natural killer cells, and myeloid populations. Air-liquid interface cultures, which preserve tumor epithelium alongside native immune and stromal cells, have enabled testing of immune checkpoint blockade responses in vitro — a capability that classical immersed organoid cultures, which strip away most immune components, simply lack.

The tumor microenvironment receives particularly careful treatment. Cancer-associated fibroblasts secrete growth factors such as hepatocyte growth factor and heparin-binding EGF-like growth factor that can shield tumor cells from targeted drugs; tumor-associated macrophages and myeloid-derived suppressor cells blunt immune attack; and stromal cells transformed by oncogenic signaling — for example, pancreatic stellate cells in pancreatic ductal adenocarcinoma — can turn chemotherapy into an unintended pro-survival signal. Organoid co-cultures now allow researchers to reconstruct these interactions deliberately, adding defined stromal or immune populations and measuring how sensitivity shifts. Organ-on-chip platforms go further, introducing controlled fluid flow, drug gradients, and spatial organization that better approximate the physiological conditions of a tumor. Spatial transcriptomics and multiplexed imaging then map where resistant niches form within the culture, connecting cell state to physical location. The result, the authors suggest, is the ability to define resistance niches — specific microenvironments in which tumor cells are actively protected from therapy.

The translational ambition of the framework is equally clear. Patient-derived organoid-based drug sensitivity testing has already been explored prospectively across gastrointestinal cancers and other tumor types, with studies reporting meaningful concordance between organoid responses and patient outcomes in progression-free survival and overall survival. Negative predictive value — the ability to identify therapies that will not work — appears particularly strong, an attribute that could spare patients futile and toxic regimens. The authors argue that standardized functional testing, pathology-based interpretation, and careful pharmacometric normalization will be essential if organoid testing is to move from exploratory studies into routine clinical decision-making, potentially supporting adaptive clinical trial designs in which treatment is matched to live drug-response data. They are equally candid about limitations: organoid establishment rates vary by tumor type, culture conditions select for certain subclones, and immune and stromal components remain incompletely reconstructed in many systems.

What emerges from the review is a deliberate repositioning of the organoid itself. The authors insist that the value of their framework lies not in replacing existing models of resistance — clonal evolution, persister biology, stem cell plasticity — but in integrating them along measurable dimensions within a clinically relevant system. PDOs, in this vision, are not merely drug screening tools but living systems for mapping the dynamic architecture of therapy resistance: an experimental arena in which genetics, cell state, cell cycle, and microenvironment can be observed interacting under longitudinal drug pressure. As co-culture methods, organ-on-chip engineering, spatial transcriptomics, and artificial intelligence-assisted image analysis continue to converge, the prospect of predicting, monitoring, and ultimately preempting the evolution of resistance in an individual patient’s tumor moves from aspiration toward feasible clinical practice. The resistance problem in cancer may not have a single answer, but it may finally have a laboratory in which all of its faces can be studied at once.

Subject of Research: Patient-derived organoids for modeling therapy tolerance, persistence, and resistance in cancer

Article Title: Patient-derived organoids for modeling therapy tolerance, persistence, and resistance in cancer

Article References: Harada, K., Sakamoto, N., Ukai, S., & Ishikawa, S. (2026). Patient-derived organoids for modeling therapy tolerance, persistence, and resistance in cancer. Molecular Cancer. https://doi.org/10.1186/s12943-026-02800-9

Image Credits: AI Generated

DOI: 10.1186/s12943-026-02800-9

Keywords: patient-derived organoids, therapy resistance, drug-tolerant persisters, cancer models, tumor microenvironment, personalized medicine, drug sensitivity testing, cancer stem cells, targeted therapy, immunotherapy, single-cell analysis, Molecular Cancer

The research centers on patient-derived organoids, or PDOs, three-dimensional cultures grown from fragments of a patient’s own tumor that preserve much of the architecture, cellular diversity, and behavior of the original cancer. Unlike long-established cell lines, which accumulate mutations and drift away from the biology of the tumors they came from, organoids are meant to serve as faithful, living stand-ins for the disease as it exists in a specific patient. That fidelity is precisely what makes them valuable for studying how treatment changes cancer over time, because any difference observed after drug exposure can be attributed to the treatment itself rather than to the artifacts of laboratory adaptation.

To build their models, the researchers collected tumor specimens from four patients with ovarian cancer who had received platinum-based chemotherapy, the backbone of first-line treatment for this disease. For one of the four cases, they went further, generating organoids not only from the primary tumor but also from paired metastatic lesions and from tumor cells floating in ascites fluid, the fluid that accumulates in the abdomen as the disease progresses. This gave them a rare opportunity to ask whether organoids derived from different sites within the same patient would retain the molecular fingerprints of their tissue of origin.

Validation came through genomic comparison. When the team sequenced the organoids and compared them against the tumor samples from which they were derived, they found a greater than 90 percent overlap in single nucleotide variants, the single-letter DNA changes that act as barcodes of a tumor’s evolutionary history. This high degree of concordance established that the organoids were genuinely representative of the cancers they came from, retaining both the primary tumor’s properties and the intertumoral heterogeneity that makes ovarian cancer so variable from patient to patient and from lesion to lesion within a single patient.

With the models validated, the researchers turned to their central question: what happens to a tumor’s drug sensitivity after it encounters chemotherapy? They exposed the organoids to a three-day pulse of the standard first-line combination of carboplatin and paclitaxel, mimicking in miniature the kind of treatment patients receive in the clinic. After this exposure, they measured how the organoids responded to a panel of therapeutic agents used in the adjuvant and recurrent settings, the drugs oncologists reach for when cancer returns or when additional consolidation therapy is needed.

The results were striking. Organoids that had been exposed to chemotherapy showed higher relative viability when rechallenged with the carboplatin and paclitaxel combination than their matched, chemo-naive counterparts, a laboratory reflection of the clinical reality that tumors become harder to kill after the first round of treatment. More intriguingly, the chemo-exposed organoids displayed a reshuffled sensitivity profile toward other drugs: they became more sensitive to cediranib, an angiokinase inhibitor that blocks the blood vessel signaling pathways tumors rely on, and more resistant to topotecan, a topoisomerase inhibitor used in recurrent disease. The treatment had not simply made the cells tougher across the board; it had selectively rewired which vulnerabilities remained open.

Perhaps the most consequential finding came when the researchers sequenced the chemo-exposed organoids and compared them to their treatment-naive counterparts. Despite the clear functional changes in drug response, the genomes remained stable. No new mutations had emerged to explain the altered sensitivity. This decoupling of phenotype from genotype carries significant implications for how resistance is understood and monitored. If a tumor can change its therapeutic profile without changing its DNA, then genomic sequencing alone, however sophisticated, cannot capture the full picture of how a patient’s cancer will respond to the next drug. The changes must instead live in other layers of biology, potentially including epigenetic modifications, alterations in gene expression, shifts in protein signaling networks, or changes in the composition of the cell populations that make up the tumor.

This phenomenon, sometimes described as non-genomic or phenotypic drug resistance, has been observed in other cancer types, but demonstrating it in patient-derived models of ovarian cancer is an important step. It suggests that the plasticity of ovarian cancer cells, their ability to shift states in response to environmental pressures like chemotherapy, may be a central driver of the recurrence and treatment failure that make this disease so lethal. It also raises the possibility that some of these treatment-induced states could be reversible, or that drugs like cediranib, to which chemo-exposed cells become more sensitive, could be strategically deployed in the window after platinum therapy when those vulnerabilities are exposed.

The technical achievement of the study lies as much in the models as in the findings. Generating organoids from ascites fluid and metastatic lesions, not just primary tumors, demonstrates that the approach can capture the full anatomical spread of the disease. The greater than 90 percent single nucleotide variant overlap between organoids and source tumors provides a quantitative benchmark for model validity that other laboratories can adopt. And the demonstration that a short, three-day chemotherapy pulse is sufficient to reveal differences in drug sensitivity suggests that these experiments can be performed quickly enough to be clinically meaningful, potentially within the timeframe of treatment decision-making.

The authors caution that the study involved a small number of patient cases, and broader cohorts will be needed to determine how generalizable these patterns are across the molecular subtypes of ovarian cancer. Still, the conceptual message is clear and potentially practice-changing: short-term culture of patient-derived organoids is sufficient to reveal differences in drug sensitivity that arise from chemotherapy exposure, and those differences can exist entirely beneath the radar of genomic testing. For a disease in which the standard of care has remained largely unchanged for decades and in which most patients eventually relapse with resistant disease, models that can expose the non-genomic dimensions of resistance offer a new starting point for designing the sequential, adaptive treatment strategies that ovarian cancer patients urgently need. The work was supported by the National Cancer Institute and the Department of Defense, and was conducted under institutional review board approval at the University of Iowa in accordance with the Declaration of Helsinki.

Subject of Research: Chemotherapy-induced, non-genomic changes in drug sensitivity studied in patient-derived ovarian cancer organoid models.

Article Title: Exposure to chemotherapy alters secondary drug sensitivity without genomic alterations in patient-derived ovarian cancer organoid models

Article References: Newtson, A. M., Uhl, D. P., Kolpin, E. S., Malmrose, P. K., Parks, S., Rush, C. M., Gabrilovich, S., Bi, J., Devor, E. J., Colling, K. E., Losh, H., Andrew-Udoh, J., Gertz, J., de la Puente, P., Leslie, K. K., & Thiel, K. W. (2026). Exposure to chemotherapy alters secondary drug sensitivity without genomic alterations in patient-derived ovarian cancer organoid models. Journal of Ovarian Research. https://doi.org/10.1186/s13048-026-02268-7

Image Credits: AI Generated

DOI: 10.1186/s13048-026-02268-7

Keywords: ovarian cancer, patient-derived organoids, chemotherapy resistance, carboplatin, paclitaxel, cediranib, topotecan, non-genomic drug resistance, platinum-based chemotherapy, personalized medicine, tumor heterogeneity, drug sensitivity testing