Júlia Vallverdú Ginés, PhD
Lead ScientistPatient-derived tumor spheroids are three-dimensional multicellular aggregates generated by dissociated tumor tissue that self-assemble in culture. Unlike conventional spheroids built from immortalized cell lines, they preserve the genetic heterogeneity of the source tumor and can also retain native stromal or tumor-infiltrating immune cells.
This description sounds close to that of a patient-derived tumor organoid, and indeed, both models are typically generated from the same starting material: a patient’s own tumor tissue. Nevertheless, organoids require conditions that selectively expand stem/progenitor-like epithelial cells into self-organizing, tissue-mimicking structures, in which process they tend to lose the stromal and immune cells. Besides, spheroids are faster to establish, allowing to accelerate timely clinical decisions.
This are not the only characteristics that sets organoids and spheroids apart in the patient-derived context:
Why Patient-Derived Tumor Spheroids Outperform Cell Line Models in Oncology Drug Screening
Conventional tumor spheroids, built from established cancer cell lines, lack genetic heterogeneity by definition, since they originate from a single clonal population. They accumulate genetic changes over years in culture, gradually diverging from the tumor they originally came from. A spheroid built from such a line can indicate whether a drug works against a cancer subtype in general, but it cannot capture how a particular patient’s tumor will actually respond.
Patient-derived spheroids, in contrast, retain the heterogeneity of the source tumor. They reflect the same patient-to-patient variability seen in the clinic and provide a useful screening platform for testing potential drugs directly against a patient’s own cancer cells before committing to a treatment decision.
For immuno-oncology applications specifically, some patient-derived spheroid formats preserve native tumor-infiltrating lymphoid and myeloid populations from the source tumor, including a T-cell receptor repertoire representative of the parent tumor, and show responsiveness to ex vivo PD-1/PD-L1 blockade. Cell-line spheroids, lacking any immune compartment, simply cannot be used to ask these questions.
Evidence for this predictive value now spans multiple cancer types and treatment modalities:
- In breast cancer, spheroids grown directly from surgical resections reproduced the ER, PR, HER2, and GATA-3 expression pattern of the source tumor, and showed viability loss following treatment with paclitaxel, 5-fluorouracil, and epirubicin (Hofmann et al. 2022).
- In ovarian cancer, spheroids grown from patient tumor tissue predicted response or non-response to first-line therapy with 89% accuracy compared to actual clinical outcomes (Shuford et al. 2019).
- In metastatic brain tumors, spheroids were tested against the same drugs each patient subsequently received, and the ex vivo response was compared directly against clinical imaging outcomes: spheroid-predicted response matched the clinical outcome in 73.3% of cases, once treatments outside the model’s scope (immunotherapy, angiogenesis inhibitors, cell therapy) were excluded (Chang et al. 2024).
- In metastatic colorectal cancer, micro-organospheres generated from digested biopsy material returned oxaliplatin sensitivity results within 14 days of biopsy, with results matching the patients’ actual clinical response to the drug (Ding et al. 2022).
Spheroid formation methods and their impact on assay reproducibility
Spheroid generation falls into two broad categories: scaffold-free methods, where cells self-aggregate, and scaffold-based methods, where dissociated tumor cells are embedded in an ECM-mimicking hydrogel.
Scaffold-free methods rely on the natural tendency of cells to aggregate. Culturing cells on low-attachment or non-adhesive surfaces is the most common approach.
Scaffold-based methods instead embed dissociated cells directly into a supportive matrix. The scaffold provides both a mechanical framework for three-dimensional growth and biochemical cues that more closely resemble the native extracellular matrix, supporting cell proliferation and differentiation. Matrix concentration is a key variable: denser formulations better mimic the stiffness of real tumor tissue, while softer, more dilute gels are more permissive for cell migration and are chosen when the application specifically calls for it. Some protocols take this a step further by loading the cell-matrix mixture into microfluidic devices, allowing controlled, low-shear culture and treatment delivery directly to the embedded spheroids.
Tissue processing adds a further layer on top of whichever method is chosen. How finely a tumor sample is minced or enzymatically digested determines the size distribution of the resulting cell clusters, and in turn how many usable spheroids can be generated from a single biopsy. Overly aggressive digestion breaks clusters down into single cells and reduces spheroid yield, while insufficient processing leaves pieces too large and heterogeneous to form uniform spheroids.
Scaling spheroid drug screening: automation and high-throughput workflows
Tumor tissue is a finite resource. A single biopsy or resection yields a fixed number of cells, which limits how many treatment conditions, doses, and replicates can be tested from any one patient sample. Once the tumor has been dissociated and the spheroid fraction isolated, seeding, medium exchange, treatment delivery, and endpoint assays are typically still performed by hand, and manual handling introduces variability.
The MO:BOT is a modular benchtop automation platform that takes over from this point, standardizing spheroid seeding, medium exchange, and treatment across a workflow of plug-and-play modules, including tilting, shaking, or cooling, that connect without cables and are recognized automatically.
Onboard quality control is handled by the MO:CROSCOPE, an on-deck imaging module that scans a full 96-well plate with per-well autofocus in around 10 minutes, segmenting every well in real time using a deep learning model trained on brightfield 3D culture morphology. Results are flagged well-by-well as pass or fail, and a plate-wide heatmap of any selected morphology metric makes it possible to spot edge effects, outlier wells, or treatment-group patterns immediately, without removing samples from the sterile environment.
(A) Typical 3D cell culture workflow phases. (B) Cell culture workflow integration in the MO:BOT. (C) Separate instruments in traditional workflows (e.g., biosafety cabinet, liquid handler, shaker, plate reader, microscope etc.), replaced by the MO:BOT. Figure adapted from Hellmold D, et al. Automation of 3D liver spheroid generation and acetaminophen dose–response on the MO:BOT enhances assay robustness and precision. Sci Rep (2026). Licensed under CC BY 4.0.
Standardizing everything downstream of tissue dissociation, and flagging low-quality spheroids in real time rather than at the end of a multi-day assay, can help turn patient-derived spheroid screening into a workflow that scales across treatment conditions and patients.
Sources
Bozym DJ, et al. Patient-derived organotypic tumor spheroids, tumoroids, and organoids: advancing immunotherapy using state-of-the-art 3D tumor model systems. Lab Chip. 2025 Jun 24;25(13):3038-3059. doi: 10.1039/d5lc00062a.
Chang CJ, et al. Preliminary study of utilizing a patient derived tumor spheroid model to augment precision therapy in metastatic brain tumors. Sci Rep. 2024 Dec 30;14(1):31888. doi: 10.1038/s41598-024-83409-0.
Ding S, et al. Patient-derived micro-organospheres enable clinical precision oncology. Cell Stem Cell. 2022 Jun 2;29(6):905-917.e6. doi: 10.1016/j.stem.2022.04.006.
Durymanov M. Tumor Spheroids, Tumor Organoids, Tumor Explants, and Tumoroids: What Are the Differences between Them? Biochemistry (Mosc). 2025 Feb;90(2):200-213. doi: 10.1134/S0006297924604234.
Hofmann S, Cohen-Harazi R, Maizels Y, Koman I. Patient-derived tumor spheroid cultures as a promising tool to assist personalized therapeutic decisions in breast cancer. Transl Cancer Res. 2022 Jan;11(1):134-147. doi: 10.21037/tcr-21-1577.
Palin CA, Jenkins RW, Revach OY. Protocol for the preparation and analysis of patient-derived organotypic tumor spheroids. STAR Protoc. 2026 Mar 20;7(1):104286. doi: 10.1016/j.xpro.2025.104286.
Shuford S, et al. Prospective Validation of an Ex Vivo, Patient-Derived 3D Spheroid Model for Response Predictions in Newly Diagnosed Ovarian Cancer. Sci Rep. 2019 Aug 1;9(1):11153. doi: 10.1038/s41598-019-47578-7.
