Organoids and AI: Unlocking the Mysteries of Cancer Immune Evasion and Immunotherapy

For decades, oncology researchers have grappled with two fundamental questions: How do cancers evade the immune system, and how can we make cancer more vulnerable to immune attack? The advent of patient-derived organoids (PDOs) combined with artificial intelligence (AI)-driven data analysis is revolutionizing the search for answers. This powerful combination is allowing scientists to see deeper into the complex interactions between cancer cells, immune cells, and normal tissues, offering unprecedented insights into tumor immunity and responses to therapy.¹

Patient-derived organoids: Expanding cancer research possibilities

These models can be co-cultured with immune cells to recreate the tumor microenvironment (TME), enabling researchers to observe how cancer cells interact with immune cells in a controlled setting (Figure 2).

Figure 2: Time-lapse images of T-cells and organoids displaying varying degrees of motility in culture and progressive reciprocal interaction. (A) Stimulated T-cells accumulating around the edge of the organoids and attracting the organoids together over time. The stimulated T-cells appear to displace organoids in culture (see white arrows).³ (B) Non-stimulated T-cells accumulating around the edge of the wells without affecting the organoids.

Commercial leaders in PDO technology have pioneered large-scale manufacturing of these organoids,⁴ allowing for high-throughput drug screening and immunotherapy testing. Such advancements make it possible to test thousands of compounds on patient-specific tumor models, accelerating the discovery of effective immunotherapies.

Challenges in creating PDOs and recent innovations

While PDOs offer immense potential, their development presents several challenges. Traditional methods of generating PDOs are labor-intensive, require specialized expertise, and can be highly variable. Additionally, the reproducibility of organoid cultures has historically been a concern, limiting their scalability for large-scale research and drug discovery.

To overcome these challenges, commercial assay-ready PDO lines and bespoke organoid solutions have been introduced. These standardized organoids are grown using advanced bioprocessing techniques, ensuring batch-to-batch consistency and eliminating the need for labor-intensive culture preparation. By providing pre-validated organoid models,⁵ researchers can focus on conducting experiments rather than on the complexities of organoid cultivation. Furthermore, tailored solutions⁶ enable scientists to develop patient-specific models that closely mimic individual tumor responses, paving the way for personalized oncology research and drug screening.

How cancers evade the immune system: Insights from organoids and AI

Cancer cells employ various mechanisms to escape immune surveillance, including immune checkpoint activation, where tumors express proteins like PD-L1 that inhibit T-cell activation;⁷ immunosuppressive TME, where cancer-associated fibroblasts and myeloid-derived suppressor cells create a hostile environment for immune cells; and metabolic barriers, in which high lactic acid production in tumors suppresses T-cell function.

AI-powered analysis is playing a crucial role in deciphering these immune evasion strategies. Using machine learning algorithms, researchers analyze vast datasets from tumor-immune organoid co-cultures. AI models can identify immune escape patterns, such as changes in T-cell activation markers and cytokine secretion profiles. In one study, AI-driven analysis of lung cancer organoids revealed pathways through which tumors suppress T-cell activation, laying the foundation for targeted interventions.⁸

Enhancing cancer vulnerability to immunotherapy

To counteract immune evasion, scientists are leveraging PDOs to test and refine immunotherapies, including checkpoint inhibitors and chimeric antigen receptor (CAR) T-cell therapy. By co-culturing patient-specific organoids with therapeutic immune cells, researchers can assess the effectiveness of immunotherapies in real-time.

Recent studies highlight how CAR T-cell therapy alters cancer organoid morphology.⁹ AI-based image analysis tools track changes in organoid shape, size, and viability over time, offering a quantitative measure of therapy effectiveness. These models allow researchers to optimize CAR T-cell designs by evaluating different antigen targets and engineering strategies.

Furthermore, high-content imaging combined with AI can predict which patients are most likely to respond to immunotherapy. For example, convolutional neural networks (CNNs) trained on PDO images can classify tumors based on their responsiveness to immune attack.^10,11 This approach could lead to personalized cancer treatments tailored to an individual’s unique tumor-immune landscape.

Cutting-edge technologies driving innovation

The integration of advanced technologies is propelling cancer research forward:

1. Cancer PDOs for drug discovery: Accessible 3D models, such as assay-ready colorectal cancer (CRC) tumoroids, reveal the subtle effects of oncotherapies on cellular viability, morphology, cytoskeleton, and mitochondria.^9,12,13
2. Automated high-throughput imaging: AI-powered tools, such as the ImageXpress HCS.ai High-Content Screening System, has full environmental control which enables real-time monitoring of immune cell infiltration into tumor organoids.¹⁴
3. Deep learning for organoid classification: AI-driven segmentation, using deep learning U-net models, accurately identify changes in organoid morphology induced by T-cell activity (Figure 3).⁹
4. Single-cell RNA sequencing: Combining organoids with single-cell sequencing reveals how individual cancer and immune cells interact, identifying potential immunotherapy targets.¹⁵

Figure 3: A. Representative TL image and B. the corresponding segmentation of intact organoids (co-cultured with non-stimulated T cells) at the first timepoint. C. Representative TL image and D. the corresponding segmentation of intact organoids at the last timepoint. E. Representative TL image and F. the corresponding segmentation of modified organoids (co-cultured with stimulated T cells) at the last timepoint. G. The representative images of classified organoids showing the increased number of modified organoids throughout the timepoint in wells treated with stimulated T cells (Pink: intact organoids, Green: modified organoids). All analyses were conducted using IN Carta Image Analysis Software. Credit: Molecular Devices.⁹

Future directions and clinical implications

The combination of organoids and AI is transforming cancer immunotherapy research. As these technologies continue to evolve, they hold the promise of enabling truly personalized cancer treatment. By predicting patient responses to immunotherapy and uncovering novel therapeutic targets, organoid-based AI models are accelerating the transition from laboratory discoveries to clinical applications.

In the coming years, integrating multi-omics data—including genomics, transcriptomics, and proteomics—with AI-driven organoid studies could lead to even more precise and effective cancer treatments. The ability to test immunotherapies in a patient-specific context before clinical application could drastically improve outcomes and reduce treatment failures.

Conclusion

Cancer research is entering a new era, where AI and PDOs are providing unprecedented insights into immune evasion and therapy resistance. By leveraging these cutting-edge technologies, scientists are not only answering some of the oldest questions in oncology but also paving the way for more effective and personalized cancer treatments. As our understanding of tumor-immune interactions deepens, the dream of harnessing the immune system to defeat cancer is becoming an increasingly tangible reality.

References

1. Thorel L, Perréard M, Florent R, et al. Patient-derived tumor organoids: a new avenue for preclinical research and precision medicine in oncology. Exp Mol Med. 2024;56:1531–1551.
2. Molecular Devices. Organoids. Available at: https://www.moleculardevices.com/applications/3d-cell-models/organoids#:~:text=Patient-derived%20tumor%20organoids%20or,and%20investigate%20cancer%20cell%20growth. Accessed February 21, 2025.
3. Tong Z, Lim A. A novel workflow to quantify the interaction between T-cells and patient-derived organoids. Molecular Devices Scientific Poster. Available at: https://www.moleculardevices.com/en/assets/scientific-posters/dd/img/novel-workflow-to-assess-t-cell-and-patient-derived-organoid-interaction. Accessed February 21, 2025.
4. Molecular Devices. Patent-pending bioprocess has the potential to produce tens of millions of quality-controlled organoids per batch. Available at: https://www.moleculardevices.com/technology/bioprocess-technology-for-organoid-production. Accessed February 21, 2025.
5. Molecular Devices. 3D Biology. Available at: https://www.moleculardevices.com/products/3d-biology. Accessed February 21, 2025.
6. Molecular Devices. 3D Ready Organoid Expansion Service. Available at: https://chat.moleculardevices.com/3dbiology-flyer-organoid-expansion-service. Accessed February 21, 2025.
7. Immune Checkpoint Inhibitors. National Cancer Institute. US National Centers of Health. Available at: https://www.cancer.gov/about-cancer/treatment/types/immunotherapy/checkpoint-inhibitors#:~:text=Checkpoint%20proteins%2C%20such%20as%20PD,the%20body%20(left%20panel). Accessed March 3, 2025.
8. Picca F, Giannotta C, Tao J, et al. From Cancer to Immune Organoids: Innovative Preclinical Models to Dissect the Crosstalk between Cancer Cells and the Tumor Microenvironment. Int J Mol Sci. 2024;25(19):10823.
9. Tong Z. T cell-induced morphological change analysis of colorectal cancer organoids using AI. Molecular Devices Application Note. Available at: https://www.moleculardevices.com/en/assets/app-note/dd/img/t-cell-induced-morphological-change-analysis-of-colorectal-cancer-organoids-using-ai. Accessed February 21, 2025.
10. Molecular Devices. IN Carta Image Analysis Software. Available at: https://www.moleculardevices.com/products/cellular-imaging-systems/high-content-analysis/in-carta-image-analysis-software. Accessed February 21, 2025.
11. Sirenko O, Roberts R. Advancements in Early Toxicity Testing. Molecular Devices Podcast. February 11, 2025. Available at: https://www.moleculardevices.com/lab-notes/cellular-imaging-systems/advancements-in-early-toxicity-testing. Accessed February 21, 2025.
12. Fraser E, Lim A. A case study for assay-ready patient-derived organoids (PDOs) and high-throughput 3D imaging to advance drug discovery. Molecular Devices. February 10, 2023. Available at: https://www.moleculardevices.com/lab-notes/3d-biology/case-study-assay-ready-patient-derived-organoids-and-high-throughput-3d-imaging. Accessed February 21, 2025.
13. Olsen C, Tong Z, Macha P, et al. A walkaway solution for assessing drug effects in patient-derived colorectal cancer organoids. Molecular Devices. Available at: https://www.moleculardevices.com/sites/default/files/en/assets/scientific-posters/br/walkaway-solution-for-assessing-drug-effects-in-patient-derived-colorectal-cancer-organoids.pdf. Accessed February 21, 2025.
14. Tong Z, Lim A. 3D quantitative measurement of T-cell penetration into cancer spheroids. Molecular Devices Application Note. Available at: https://www.moleculardevices.com/en/assets/app-note/dd/img/3d-quantitative-measurement-of-the-t-cell-penetration-into-cancer-spheroids. Accessed February 21, 2025.
15. Verduin M, Hoeben A, De Ruysscher D, Vooijs M. Patient-Derived Cancer Organoids as Predictors of Treatment Response. Front Oncol. 2021:11:641980.

Recent posts