Convergent bioprinting and microfluidics: toward next-generation biomimetic tumor models

Precision medicine continues to evolve, and the complex nature of tumor biology drives the need for in vitro models that are both highly biomimetic and functionally integrated. Conventional two-dimensional cell cultures cannot adequately reproduce the three-dimensional (3D) architecture and microenvironmental diversity of tumors. Animal models, meanwhile, face constraints such as species differences, limited throughput, and difficulties in real-time observation. Transformative in nature, the convergence of 3D bioprinting and microfluidic technologies now allows the creation of advanced platforms that mirror tumor spatial organization, cell-cell communication, and dynamic physiological contexts. This combined approach does more than enhance our basic grasp of tumor biology—it also offers considerable translational promise for personalized drug testing, immunotherapy assessment, and exploring how metastasis occurs.

This article undertakes a thorough examination of noteworthy progress in combining 3D-bioprinted tumor models with microfluidic platforms. It outlines the trajectory of this interdisciplinary field—evolving from static 3D forms to dynamic, life-like systems; shifting focus from modeling isolated tumors to constructing intricate vascularized and immunocompetent niches; and advancing from manual handling to automated, intelligent analysis. Additionally, the discussion identifies pressing interdisciplinary hurdles and contemplates future pathways for deploying such integrated models within precision oncology.

While conventional 3D bioprinting produces tumor models with precise spatial architectures—spheroids and organoids being prime examples—these constructs frequently lack essential dynamic physiological signals, such as nutrient exchange, fluid shear stress, and metabolic gradients. Consequently, such static models regularly prove inadequate for faithfully mimicking the actual growth and invasive characteristics of tumors observed in vivo (1,2). Microfluidic technology, on the other hand, permits exact microscale fluid control, supporting perfusion culture, tailored mechanical stimulation, and customizable chemical gradients. This functionality directly addresses the shortcomings inherent in static 3D cultures (3,4). Merging bioprinting with microfluidics therefore creates opportunities to develop vascularized, perfusable, and mechanically adaptive living tumor microenvironments on chip-based platforms.

In one illustrative example utilizing an osteosarcoma-on-a-chip model (1), researchers combined 3D-bioprinted heterotypic tumor-stromal constructs with a microfluidic perfusion system to examine how hemodynamic shear stress influences malignant phenotypes and drug sensitivity. The dynamic culture conditions not only increased tumor cell invasiveness but also improved chemosensitivity, a conclusion supported by markedly lower half-maximal inhibitory concentration (IC₅₀) values relative to static controls. This evidence underscores how biomechanical signals within the tumor microenvironment can crucially shape the prediction of therapeutic outcomes.

Bioinks function as the essential “living materials” in 3D bioprinting, directly governing the viability, functionality, and biomimetic fidelity of the resulting constructs. By adjusting the concentration and composition of bioinks, extracellular matrix (ECM) of varying degrees of softness and hardness can be simulated to construct biomimetic in vitro models of diseases such as liver cirrhosis and fatty liver (5,6). To illustrate, a GelMA-gelatin-fibrin composite bioink designed for breast cancer models successfully supported cell proliferation while also encouraging microvascular network formation (2). In a strategy applied to bone tumors, hydroxyapatite was blended into GelMA to replicate the mineralized bone ECM, allowing tissue-specific simulation (1). Moving forward, research is progressing toward bioinks derived from tissue-specific decellularized ECM (dECM) and the inclusion of smart responsive materials, such as thermosensitive or pH-responsive polymers. These advanced inks are anticipated to more dynamically emulate the changing tumor microenvironment throughout disease development (7).

Within tumors, the vascular system acts as a vital pathway for nutrient supply and metastatic dissemination, yet it also constitutes a significant physiological barrier that hinders effective drug delivery. To replicate this complex interface, microfluidic strategies have been employed to create perfusable 3D microvascular networks on chip platforms, which are then integrated with tumor spheroids or organoids (8-10). A prominent illustration is the “hTPV-Chip” (9), where a co-culture of patient-derived tumor organoids and self-assembled vascular networks was implemented. This configuration allowed real-time tracking of highly metastatic tumor cells as they migrated directionally along blood vessels, and revealed the Notch signaling pathway as a central regulator of this process. Vascularized organoid-chip models of this kind provide a physiologically meaningful platform to study tumor-vascular interactions, uncover mechanisms of metastasis, and evaluate anti-angiogenic drugs.

Conventional drug screening frequently faces limitations such as low throughput, substantial reagent use, and cumbersome manual steps. Digital microfluidics (DMF) and active-matrix digital microfluidics (AM-DMF) offer a solution through precise, parallel handling of nanoliter-scale droplets (11-13). As an example, the AM-DMF platform permits fully automated droplet-based dispensing of patient-derived organoids, programmable drug gradient generation, and longitudinal fluorescence imaging across 72 hours. Given its low cell-number requirement, the platform is especially appropriate for personalized drug testing when only small clinical biopsy samples are available. Moreover, coupling such a system with deep learning-based image analysis—for instance, YOLOv8—enables automatic identification and sorting of target cells or organoids (11), moving the field closer to intelligent, on-demand experimental systems.

Resolving tumor heterogeneity requires analytical methods capable of single-cell resolution. Models generated by 3D bioprinting and microfluidics provide spatially defined or phenotypically segregated cell populations, and contemporary microfluidic systems are now more frequently combined with single-cell proteomic workflows (10,14,15). One case is the DMF-Try@Fe₃O₄ platform (14), where nanoscale-immobilized enzyme reactors reduce the proteolytic digestion time for individual cells from several hours to merely 10 minutes, markedly increasing throughput and proteome coverage. Likewise, the PISPA platform (15) attains deep quantification of as many as 3,000 proteins per cell by employing probe-based sorting and nanoliter-scale reaction chambers. When linked with tumor-on-a-chip models, these techniques help form a closed research loop that spans model construction, phenotypic tracking, and final molecular profiling at the single-cell level.

Modern tumor modeling has progressed from static 3D forms to systems that embed dynamic physiological cues—such as fluid shear stress, mechanical loading, and cyclic perfusion—thereby more faithfully mimicking tumor growth, invasion, and drug response (1,3,4). Earlier platforms often contained only tumor cells; present designs, in contrast, draw stronger inspiration from biology, incorporating varied stromal cells (fibroblasts, endothelial cells), immune constituents, neural inputs, and sophisticated ECM compositions (2,7,8). Models personalized with patient-derived tumor cells, organoids, or stromal subsets are increasingly applied to forecast clinical outcomes, supporting the progress of precision oncology (9,13,16). The incorporation of high-resolution live-cell imaging, biosensors, and online analytical tools like mass spectrometry further allows prolonged, multiparametric tracking of cellular behaviour, metabolic dynamics, and spatio-temporal drug distribution in these models (9,10,17). A final notable shift is toward automation and intelligence: through digital DMF, robotic platforms, and artificial intelligence (AI)-powered image analysis, fully automated, reproducible, high-throughput workflows now encompass model fabrication, culture, stimulation, and detection (11-13).

Integrated tumor models exhibit considerable translational promise in diverse clinical and preclinical contexts. As exemplified by a 3D-bioprinted gastric cancer model produced within ten days, drug sensitivity results aligned closely with actual patient chemotherapy responses, underscoring its clinical relevance for informing perioperative treatment plans (16). In situations where tissue is scarce, the AM-DMF organoid platform supports high-throughput drug testing starting from minimally invasive biopsies (13). Within immunotherapy research, vascularized tumor-chip systems have successfully modeled the entire CAR-T cell cascade—covering vascular perfusion, extravasation, tumor infiltration, and cytotoxic killing—which facilitates assessment of engineered CAR-T cells (e.g., CCR2-modified variants) and pinpoints synergistic targets like DPP4 inhibitors (17). These sophisticated in vitro models are becoming indispensable for addressing the predictive shortcomings of animal models in solid-tumor CAR-T therapy (18). Additionally, this model holds significant potential in the field of immunotherapy research for liver cancer (19), particularly when integrated with single-cell transcriptomics to elucidate the immune heterogeneity within the tumor microenvironment (20). Moreover, vascularized constructs (8,9) and multi-organ chips that simulate metastatic cascades (4,7) furnish distinctive platforms for deconstructing pivotal metastatic steps, ranging from vascular/lymphatic invasion and systemic circulation to distant colonization, thereby speeding the development of new anti-metastatic treatments.

Notwithstanding considerable progress, combining 3D bioprinting with microfluidic tumor models confronts enduring hurdles. Present systems often omit key physiological elements—functional immune niches, neural inputs, and systemic multi-organ interplay. Achieving reproducible results also poses a barrier, stemming from batch-dependent bioink variability, complex chip fabrication steps, and an absence of uniform laboratory protocols. Moreover, tightly interfacing these models with multi-omics analytics—spatial transcriptomics, metabolomics, proteomics—requires deeper technological convergence. A further constraint is the substantial expense of advanced bioprinters and automated microfluidic setups like AM-DMF platforms, which still restricts their widespread use in everyday research and clinical practice.

Looking forward, a number of strategic pathways are set to propel the field forward. Key directions involve creating multi-organ-on-a-chip systems that can simulate systemic processes like tumor metastasis and off-target drug effects; moving toward “patient-on-a-chip” platforms which bring together patient-specific tumors, immune and stromal cells, and genetic context to tailor disease modeling and therapy prediction; and expanding the use of AI and machine learning to refine bioprinting parameters, devise advanced chip architectures, and decipher intricate spatiotemporal omics data. Strengthening collaboration across disciplines—materials science, engineering, cell biology, and clinical research—will ultimately be crucial for setting rigorous model validation benchmarks and well-defined translational routes.

The combination of 3D bioprinting and microfluidics is shifting in vitro tumor modeling from static constructs toward refined, next-generation tumor-on-a-chip systems. Such integrated platforms deliver dynamic physiological fidelity, patient-specific design, and high-throughput analysis. This cross-disciplinary synergy affords not only a clearer window into the intricacies of tumor biology, but also powerful new tools for precision drug screening, personalized therapy prediction, and novel treatment discovery. By tackling remaining obstacles through continued interdisciplinary innovation and cooperation, these highly biomimetic models offer real potential to shorten the path from laboratory to clinic, and in doing so, help improve outcomes for cancer patients.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, HepatoBiliary Surgery and Nutrition. The article did not undergo external peer review.

Funding: This work was supported by the National Key R&D Program of China (No. 2024YFB4607800) and Peking Union Medical College Hospital Talent Cultivation Program (Category C) (No. UBJ10452).

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2026-1-0021/coif). H.Y. serves as an unpaid editorial board member of HepatoBiliary Surgery and Nutrition. The other author has no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Jaiswal C, Dey S, Prasad J, et al. 3D bioprinted microfluidic based osteosarcoma-on-a chip model as a physiomimetic pre-clinical drug testing platform for anti-cancer drugs. Biomaterials 2025;320:123267. [Crossref] [PubMed]
2. Yuan T, Fu X, Hu R, et al. Bioprinted, spatially defined breast tumor microenvironment models of intratumoral heterogeneity and drug resistance. Trends Biotechnol 2024;42:1523-50. [Crossref] [PubMed]
3. Shao C, Yu Y, Lei X, et al. Organ-on-a-chip for dynamic tumor drug resistance investigation. Chemical Engineering Journal 2023;460:141739.
4. Liu X, Fang J, Huang S, et al. Tumor-on-a-chip: from bioinspired design to biomedical application. Microsyst Nanoeng 2021;7:50. [Crossref] [PubMed]
5. Yan Y, Zhang K, Li F, et al. The gut-liver axis links the associations between serum carotenoids and non-alcoholic fatty liver in a 7.8-year prospective study. Hepatobiliary Surg Nutr 2025;14:16-32. [Crossref] [PubMed]
6. Hsu HC, Chow LH, Chen YL, et al. Effects of exercise and nutrition in improving sarcopenia in liver cirrhosis patients: a systematic review and meta-analysis. Hepatobiliary Surg Nutr 2025;14:33-48. [Crossref] [PubMed]
7. Monteiro CF, Deus IA, Custódio CA, et al. Biomaterials meet organ-on-chips - a perspective on tumor modeling. International Materials Reviews 2024;70:31-68.
8. Abreu CM, Lima AC, Neves NM, et al. MicroVasculoid-Chip: A 3D Self-Assembled Human Microcirculation-on-a-Chip Model Reveals Enhanced Lymphangiogenic Lung Cancer-Induced Vessel Remodeling and Invasion. Advanced Materials Technologies 2025;10:2400883.
9. Du Y, Wang YR, Bao QY, et al. Personalized Vascularized Tumor Organoid-on-a-Chip for Tumor Metastasis and Therapeutic Targeting Assessment. Adv Mater 2025;37:e2412815. [Crossref] [PubMed]
10. Li H, Ma Y, Fu R, et al. Droplet-Based Microfluidics with Mass Spectrometry for Microproteomics. Engineering 2024;43:37-53.
11. Guo Z, Li F, Li H, et al. Deep Learning-Assisted Label-Free Parallel Cell Sorting with Digital Microfluidics. Adv Sci (Weinh) 2025;12:e2408353. [Crossref] [PubMed]
12. Hu C, Chang C, Zhang M, et al. "Cell-On-Demand" Digital Microfluidics for Real-Time Low-Abundance Single-Cell Isolation and Sample Analysis. Small 2025;21:e2504239. [Crossref] [PubMed]
13. Dong W, Sun R, Feng Z, et al. Functional drug screening of tumor organoids on an Active-Matrix Digital Microfluidic Chip for cancer precision medicine. 2025. Available online: https://www.researchsquare.com/article/rs-7772352/v1
14. Zhao M, Li H, Guo Z, et al. Rapid Single-Cell Proteomics Using Nanoconfined Enzyme Reactors on a Microscale Digital Microfluidics Platform. Advanced Functional Materials 2026;36:e02142.
15. Wang Y, Guan ZY, Shi SW, et al. Pick-up single-cell proteomic analysis for quantifying up to 3000 proteins in a Mammalian cell. Nat Commun 2024;15:1279. [Crossref] [PubMed]
16. Du L, Zheng Z, Zhang K, et al. Exploring personalized prediction of clinical chemotherapy efficacy and revealing tumor heterogeneity using patient-derived 3D bioprinting gastric cancer models. Mol Cancer 2025;24:259. [Crossref] [PubMed]
17. Liu H, Noguera-Ortega E, Dong X, et al. A tumor-on-a-chip for in vitro study of CAR-T cell immunotherapy in solid tumors. Nat Biotechnol 2025; Epub ahead of print. [Crossref]
18. Sun X, Zhang Z, Zhao W, et al. Prediction of CAR-T Therapy In Vitro: Development of Biomimetic Models for CAR-T Killing Effect Assessment. Small 2025;21:e2503384. [Crossref] [PubMed]
19. Wang H, Qian YW, Dong H, et al. Pathologic assessment of hepatocellular carcinoma in the era of immunotherapy: a narrative review. Hepatobiliary Surg Nutr 2024;13:472-93. [Crossref] [PubMed]
20. Li Y, Xun Z, Long J, et al. Immunosuppression and phenotypic plasticity in an atlas of human hepatocholangiocarcinoma. Hepatobiliary Surg Nutr 2024;13:586-603. [Crossref] [PubMed]