Patient-derived organoids (PDOs) are a powerful tool in translational cancer research, reflecting both the genetic and phenotypic heterogeneity of the disease and response to personalized anti-cancer therapies. Here, a consolidated protocol to generate human primary bladder cancer PDOs in preparation for the evaluation of phenotypic analyses and drug responses is detailed.
Current in vitro therapeutic testing platforms lack relevance to tumor pathophysiology, typically employing cancer cell lines established as two-dimensional (2D) cultures on tissue culture plastic. There is a critical need for more representative models of tumor complexity that can accurately predict therapeutic response and sensitivity. The development of three-dimensional (3D) ex vivo culture of patient-derived organoids (PDOs), derived from fresh tumor tissues, aims to address these shortcomings. Organoid cultures can be used as tumor surrogates in parallel to routine clinical management to inform therapeutic decisions by identifying potential effective interventions and indicating therapies that may be futile. Here, this procedure aims to describe strategies and a detailed step-by-step protocol to establish bladder cancer PDOs from fresh, viable clinical tissue. Our well-established, optimized protocols are practical to set up 3D cultures for experiments using limited and diverse starting material directly from patients or patient-derived xenograft (PDX) tumor material. This procedure can also be employed by most laboratories equipped with standard tissue culture equipment. The organoids generated using this protocol can be used as ex vivo surrogates to understand both the molecular mechanisms underpinning urological cancer pathology and to evaluate treatments to inform clinical management.
Bladder cancer is the most prevalent urinary tract cancer and the tenth most common human malignancy worldwide1. It encompasses a genetically diverse and phenotypically complex spectrum of disease2. Urothelial non-muscle-invasive forms of bladder cancer (NMIBC) are the most common bladder cancer diagnoses (70%-80%), and these cancers display considerable biological heterogeneity and variable clinical outcomes2,3,4. Patients with NMIBC typically experience a high risk of disease recurrence (50-70%) and one-third of cancers will progress and develop into significantly more aggressive muscle-invasive bladder cancer (MIBC)2. Although 5-year survival rates for NMIBC are high (>90%), these patients must undergo long-term clinical management5. On the other hand, locally advanced (unresectable) or metastatic MIBC is generally considered incurable6. Consequently, bladder cancer has one of the highest lifetime treatment costs within cancer care and is a significant burden for both the individual and the healthcare system3,7. The underlying genetic aberrations in advanced disease renders therapeutic management of bladder cancer a clinical challenge, and therapeutic options for invasive urothelial tumors have only recently improved since the approval of immunotherapies for both advanced and high-risk NMIBC8,9. Currently, clinical decision-making has been guided by conventional clinical and histopathological features, despite individual bladder cancer tumors showing large differences in disease aggressiveness and response to therapy10. There is an urgent need to accelerate research into clinically useful models to improve the prediction of individual patient prognosis and identification of effective treatments.
Three-dimensional (3D) organoids show great potential as tumor models due to their ability to self-organize and recapitulate the original tumor's intrinsic in vivo architecture and pharmacogenomic profile, and their capability to mirror the native cellular functionality of the original tissue from which they were derived11,12,13. Although established bladder cancer cell lines are readily available, relatively cost-effective, scalable, and simple to manipulate, the in vitro cell lines largely fail to mimic the spectrum of diverse genetic and epigenetic alterations observed in clinical bladder cancers12,14 and were all established and maintained under 2D, adherent culture conditions. Additionally, cell lines derived from primary and metastatic bladder tumors harbor significant genetic divergence from the original tumor material. 8,15.
An alternative approach is to use genetically engineered and carcinogen-induced mouse models. However, while these models recapitulate some of the natural oncogenic cascades involved in human neoplasia (reviewed in refs16,17,18), they lack tumor heterogeneity, are expensive, poorly represent invasive and metastatic bladder cancer, and are not viable for rapid term drug testing as tumors can take many months to develop14,19. Patient-derived models of cancer (including organoids, conditionally reprogrammed primary cell culture, and xenografts) provide invaluable opportunities to understand the effects of drug treatment before clinical treatment20. Despite this, few groups routinely use these patient-proximal models due to limited access to fresh primary patient tissue and the extensive optimization required to reproducibly generate patient-derived organoid (PDO) culture conditions. In an in vivo setting, oncogenic cells can interact and communicate with various compositions of the surrounding constituents, including stromal cells, tissue infiltrating immune cells, and matrix12. Similarly, for PDOs grown in a 3D format, cellular/matrix complexity can be customized to include other relevant components. PDOs can be rapidly generated and are often able to be passaged extensively or cryopreserved for later use, despite having a finite lifespan21,22,23. Pharmacodynamics (i.e., response to a drug) can be evaluated using multiple read-outs, including organoid viability and morphology, and characterization of immunohistochemistry targets or transcriptional changes.
Here, the procedures for the establishment of bladder cancer organoids from patient material collected from transurethral resection of bladder tumor (TURBT) or surgical removal of the bladder (radical cystectomy) are described. The method to generate PDOs is illustrated, using readily available wet laboratory materials and tools. Endpoints include changes in cell morphological characteristics and viability. These were measured using fluorescence microscopy, in vitro viability (metabolic and cell membrane integrity) assays, and histopathological analysis. Figure 1 shows the workflow for establishing human bladder cancer PDOs from clinical material obtained during elective surgery.
Patients have consented to this study following their admission under the Urology team at the Princess Alexandra Hospital, Brisbane, Australia. This study was performed in accordance with the principles of the Declaration of Helsinki and within ethical and institutional guidelines (ethics number HREC/05/QPAH/95, QUT 1000001165).
NOTE: As eligibility criteria, patients were aged ≥ 18 with cancer, and able to understand and provide consent. Those who were not able to give informed consent were excluded. Those having a primary language other than English were excluded as the provision of interpreters was not possible due to logistical and budgetary considerations. Also excluded were patients whose tumors were not accessible to biopsy or unlikely to be available in adequate amount after routine pathology.
1. Organoid Medium Preparation
NOTE: Human bladder cancer organoid medium requires growth factors that aid in the survival, growth, and continuous expansion of organoids derived from dissociated clinical material (Table 1). For complete details of each supplement used in this procedure, please refer to the Table of Materials.
2. A day before procedure outlined in 3
3. Generation of bladder tumor organoids
NOTE: This is an initial step for PDO establishment from primary patient tumors. This procedure is adapted for bladder cancer tissues from methods established by Gao et al.24.
3D organoids were successfully established from human bladder cancer patient TURBT and cystectomy tissues. Briefly, this technique highlights a rapid formation of 3D multicellular structures that are both viable and suitable for other endpoint analyses such as histological evaluation, molecular characterization (by immunohistochemistry or quantitative real-time PCR), and drug screening. During the procedure (Figure 1), the various eluates during our filtration phases (Figure 1, steps 1–6) could be taken advantage of to cryopreserve single cells for isolation of tumor-infiltrating immune cells (TILs) and generation of a single cell biobank. The procedure allows for multiple aspects of a heterogeneous primary tumor to be biobanked appropriately and in a way that can be tracked across patients (Figure 2A,B), including histology (Figure 2C,D) and fresh frozen samples (Figure 2E).
A 2 h digestion with commercially available collagenase/hyaluronidase and DNase 1 allows for sufficient digestion of 0.5-1 mm3 pieces of the more complex cystectomy biopsy tissue, including the neoplastic cells within the lamina propria and muscular layers of the bladder (Figure 3A). Larger post-digestion fragments (>100 microns) have proven suitable for culture in standard tissue culture plates of any size for the isolation of novel 2D cultures (Figure 3B). However, extensive molecular characterization is required to validate the cancer origin of such cell lines. Organoids which are successfully generated with this protocol undergo processes of dynamic self-assembly, including phases of aggregation, compaction, and final formation, whereby they form into tight cellular structures (Figure 3C) that can be assessed for response to standard-of-care therapies. Figure 3C highlights the efficient self-assembly properties of the organoids. Histologically, these structures recapitulate the original patient tumor (Figure 4). Organoids generated in this procedure are suitable for a multitude of purposes, including further characterization, biobanking, and drug efficacy testing (Figure 5).
Figure 1: Schematic representation of organoid tissue culture from the collection of tumor samples from surgery to endpoint analyses. URN, unit record number; TIL, Tumor-infiltrating lymphocyte. (1) Tissue is mechanically minced in a 90 mm petri dish as finely as possible with a sterile scalpel blade (0.5-1 mm3 pieces). (2) Tumor pieces are resuspended in an enzymatic dissociation medium consisting of the organoid medium, collagenase/hyaluronidase, and DNase 1. (3) Minced tumor tissue and enzyme solution are incubated for 1-2 h on a rotator incubator (150 rpm, 37 °C, 5% CO2) and subsequently pelleted (4) to dissociate fragments into a finer cell suspension. The sample undergoes filtration through pre-wet reversible (5) 100 µm and (6) 37 µm strainers into new 50 mL tubes to remove large insoluble material and isolate 37-100 µm clusters. Following (7) centrifugation of the fraction isolated from the reverse of the 37 µm strainer, the cell clusters are suspended in (8) BME and organoid media before (9) live-cell imaging. Created with BioRender.com. Please click here to view a larger version of this figure.
Figure 2: Sterile tumor tissue handling for organoid establishment. (A) Upon excision and transport to the laboratory, the specimen is placed in a sterile petri dish, weighed, and subsequently dissected. (B) A specimen dissection grid is used to mark the specimen for various processes. An example of dissection is shown in (C) where samples are taken from the tumor periphery as much as possible to avoid areas of central necrosis and annotated (inset) before (D) grossing the tissue prior to tissue fixation. Scale bar: 6 mm. Inset scale bar: 4 mm. (E) Small fresh fragments are taken for subsequent freezing and biobanking. Scale bar: 5 mm. Elements of this image were created with BioRender.com. Please click here to view a larger version of this figure.
Figure 3: Enzymatic derivation and dynamic assembly of bladder tumor organoids. (A) Representative pictures of hematoxylin and eosin (H&E) stained parental bladder cancer tissue (pathology reviewed as T3bN0) on immediate fixation (left) and 2 h post-digestion with 1x collagenase/ hyaluronidase (right). Scale bar: 500 µm. Inset scale bar: 50 µm. (B) Representative images of the >100 µm, 37-1 and 70-1 (day 2) isolations. Scale bars are indicated within the image. (C) Representative images are shown for bladder cancer organoid isolation over 48 h and day 7 organoid H&E staining. Scale bar: 50 µm. Scale bar for the image in the inset: 20 µm. DIC: differential interference contrast. Elements of this image were created with BioRender.com Please click here to view a larger version of this figure.
Figure 4: Bladder cancer-derived organoid cultures maintain the histological architecture of parental tumors. Representative H&E image of low-grade bladder cancer (Ta) and corresponding bright-field image of organoid at day 7 of culture. The far-right panel shows associated H&E staining of the day 7 organoid. Scale bars are indicated in the figure. LG: low grade, BC: bladder cancer, H&E: haematoxylin and eosin, and DIC: differential interference contrast. Please click here to view a larger version of this figure.
Figure 5: Downstream applications with bladder cancer patient-derived organoids. Created with BioRender.com. Please click here to view a larger version of this figure.
Additive | Final Conc. | Stock Conc. | Solvent | |
Basal Media | Glutamax | 2 mM | 1 M | NaCl |
HEPES | 10 mM | 1 M | NaCl/NaHPO4 buffered solution | |
Supplements for complete media | R-spondin 1 conditioned media | 5% v/v | Conditioned Media | Advanced DMEM/F-12 |
Noggin conditioned media | 5% v/v | Conditioned Media | Advanced DMEM/F-12 | |
EGF | 50 ng/mL | 0.5 mg/mL | PBS/0.1% BSA | |
FGF-10 | 20 ng/mL | 100 µg/mL | PBS/0.1% BSA | |
FGF-2 | 5 ng/mL | 50 µg/mL | PBS/0.1% BSA | |
Nicotinamide | 10 mM | 1 M | 1 g in 8.2 mL ddH20 | |
N-acetyl-L-cysteine | 1.25 mM | 500 mM | 40 mL ddH20 | |
A 83-01 | 0.5 µM | 50 mM | DMSO | |
SB202190 | 10 µM | 50 mM | DMSO | |
Y27632 | 10 µM | 100 mM | ddH20 | |
B-27 additive | 1X | 50x | ||
Prostaglandin E2 | 1 µM | 10 mM | DMSO | |
Primocin | 100 µg/mL | 50mg/mL |
Table 1: Supplements used for basal and complete organoid media
Supplemental File: Please click here to download this File.
While 3D organoid protocols derived from bladder cancer tissue are still in their infancy, they are an area of active research and clinical investigation. Here, an optimized protocol to successfully establish bladder cancer PDOs that are suitable for both NMIBC and MIBC specimens is detailed. This workflow integrates parallelly into hospital-based clinical trials and considers biobank sample accrual, including histological sample processing and fresh frozen tissue banking, which is an important consideration for clinical organoid pipelines. This protocol builds on existing methods and provides a consolidated methodology to build an organoid precision medicine pipeline.
The current success rate of this protocol – like others – is approximately 70%25. It is anticipated that this will improve as isolation techniques and characterization of the bladder cancer tumor microenvironment evolves. As expected, the quality of the initial specimen influences the success rate and is an important limiting step. Specimens obtained from chemotherapy naïve patients provide the highest success rate, while those from patients who have received neoadjuvant chemotherapy may have a reduced number of viable cells26,27. Typically, bladder cancer specimens (from both TURBT and cystectomies) provide an ample amount of samples with high tumor cellularity. This allows robust generation of organoids ranging between 30-100 µm within a 24-72 h growth period (given some <40 µm clusters are retained in the filtration steps). Importantly, the heterogeneity observed in organoid samples (i.e., cystic formations and more dense structures) indicates that this protocol provides conditions suitable to derive heterogeneous tumor cell populations.
Our optimized protocol can be performed on limited starting material (as is the case from some elective surgeries, including TURBT procedures or specimens acquired from patients on a clinical trial), enabling the incorporation of multiple analyses to maximize specimen value. This protocol includes methods to describe steps to grow 2D culture from extraneous tumor cells of interest that may reside in large insoluble tissues processed during the initial filtration step. Like other research groups which suggest processing specimens within 24 h post-surgery28,29, it is observed that tissue hypoxia that occurs within this timeframe does not preclude generating viable organoids. Optimization is required to determine the density of cell (native clusters) seeding per well. In our experience, excessive numbers of clusters can result in poor growth of the cultures, while low visual density cultures can lead to cell death. This can be determined by the user empirically in the context of their volumes which can be diluted within a range as necessary. Additionally, TURBT specimens often contain burnt edges due to the resection process, which are apparent as prominent necrotic regions. A critical step is the careful isolation of macroscopic tumor tissue and subsequent dissection. This is paramount to deriving viable cultures and needs to be communicated with the surgical team. In some instances, this may be an initial limitation to the procedure.
An additional limitation to the current technique is the absence of the tumor microenvironment parallel to the organoids. A necessary development will be future co-culture systems to increase cellular complexity and provide other components of the tumor microenvironment (i.e., stroma and immune cells). As organoids are defined as self-organizing structures with the ability to self-perpetuate their growth, further work will be required to elucidate the discrete cancer stem-cells (such as CD44 and CD49) which may allow for improved in vitro growth characteristics31,32,33. Establishing with certainty that the PDOs originate from tumor tissue is critical, as the contribution from normal urothelial cells in functional and molecular analyses will confound investigations into drug resistance. Sometimes the tumor origin of the urothelial cells in the PDOs is clear due to the use of metastatic tissue or papillary tumors (as is the case here; see Figure 4), but when the source tissue is excised from the bladder mucosa and thus contains margins of the benign urothelium, it is critical to confirm that the resultant PDOs are comprised of tumor cells using molecular techniques. Thus, immunohistochemical and genomic comparisons are required between the established organoids and the primary tumor from which they were derived30.
Regarding the number and size distributions of the organoids per well, it is difficult to maintain uniform plating density of 3D organoids for cytotoxicity analysis. This is slightly mitigated by isolating defined ranges using the filters, but it is still a limitation if small clusters and larger organoids are isolated in the same well. Large particle sorters may mitigate this issue but are not widely available in medical research laboratories. Nonetheless, techniques that measure viability signals before and after treatment have been used successfully in bladder cancer organoids34. As this technology advances, this will enable researchers to develop a library of bladder cancer PDOs with varying genomic profiles, and pharmacodynamic responses that can be explored to investigate novel therapies. Additionally, the specimen collection procedure and dissociation method described herein allow concurrent preparations for tumor spatial profiling, multiparametric flow cytometry, and single-cell RNA sequencing. Taken together, this allows for a diverse range of powerful endpoints which can be brought together to explore a patient's tumor heterogeneity and applicability of drugs such as immunotherapies.
In summary, we describe a protocol to establish and evaluate cultures of 3D patient-derived bladder cancer organoids in the advent of precision medicine endpoints. It is important to note that the steps and notes described here are routine approaches used to isolate bladder cancer PDOs in our laboratory. The culture medium used in our procedure was originally described for prostate cancer cells and has now been proven to be suitable for the culture of bladder cancer. Importantly, this protocol is centralized, suitable for routine use in laboratories, and broadly applicable across urological cancer types. In combination with high fidelity molecular phenotyping and drug response endpoints, this technique will provide a useful tool to rapidly explore the precise therapeutic management of bladder cancer patients.
The authors have nothing to disclose.
We acknowledge the technical assistance of the Translational Research Institute Histology core and Biological Resource Facility. This research was supported by funding from a Princess Alexandra Research Foundation award (I.V., E.D.W.), and the Medical Research Future Fund (MRFF) Rapid Applied Research Translation Program (Centre for Personalised Analysis of Cancers (CPAC; E.D.W., I.V.). The Translational Research Institute receives support from the Australian Government.
1.2 mL cryogenic vial | Corning | 430487 | |
1.5 mL Eppendorf tubes | Sigma-Aldrich | T9661-500EA | polypropylene single-use tube |
100 µM reversible strainer | STEMCELL Technologies | #27270 | |
100 mm petri dish | Corning | 430167 | |
10x Collagenase/ Hyaluronidase | STEMCELL Technologies | #07912 | |
37 µM reversible strainer | STEMCELL Technologies | #27250 | |
37°C incubator | |||
37°C water bath | |||
50 mL falcon tube | Corning | CLS430829-500EA | |
6-well plate | Corning | CLS3516 | |
70% (w/w) ethanol | |||
-80°C freezer | |||
96 well ultra low attachment plate (Black) | Sigma-Aldrich | CLS3474-24EA | |
A 83-01 | BioScientific | 2939 | Prevents the growth-inhibitory effects of TGF-β |
ACK lysis buffer | STEMCELL Technologies | #07850 | |
adDMEM/F-12 | Thermo Fisher Scientific | 12634028 | Base medium |
Animal-free recombinant EGF | Sigma | 518179 | Growth factor |
Automated cell counter (TC20) | Bio-rad | 1450102 | |
B27 additive | Gibco | 17504044 | Increases sphere-forming efficiency |
Cell Counting Slides | Bio-rad | 1450015 | |
Centrifuge | Eppendorf | EP022628188 | |
Computer system | |||
CryoStor CS10 | STEMCELL Technologies | #07930 | Cell freezing solution |
Dispase II, powder | Thermofisher | 17105041 | To enzymatically disrupt Matrigel |
DNAse 1 | STEMCELL Technologies | #07900 | |
DPBS | Thermofisher | 14190144 | |
Dry and wet ice | |||
Esky | |||
Farmdyne (Iodine 16g/L) | Ecolab | ||
Formalin solution, neutral buffered, 10% | Sigma | HT501128 | Histological tissue fixative |
Glutamax (L-alanine-L-glutamine) | Invitrogen | 35050061 | Source of nitrogen for the synthesis of proteins, nucleic acids |
HEPES | Gibco | 15630-080 | All-purpose buffer |
Histology cassette | ProSciTech | RCH44-W | |
Human FGF-10 | Peprotech | 100-26-25 | Growth factor |
Human FGF-2 | Peprotech | 100-18B-50 | Growth factor |
Liquid nitrogen | |||
Matrigel (Growth Factor Reduced (GFR), phenol red-free, LDEV free) | In Vitro Technologies | 356231 | Basement membrane extract (BME) |
Mr. Frosty freezing container | ThermoFisher | 5100-0001 | cell freezing container |
N-acetyl-L-cysteine (NAC) | Sigma | A7250 | Anti-oxidant required to protect against ROS-induced cytotoxicity |
Nicotinamide | Sigma | N0636-100G | SIRT-1 inhibitor |
NikonTs2U inverted microscope | Nikon | MFA510BB | |
NIS-Elements Advanced Research | Nikon | MQS31000 | |
Noggin conditioned media | In-house | BMP inhibitor | |
Pipetboy acu 2 | Integra | 155000 | |
Pipettes (p20, p100, p1000) with tips | |||
Primocin | Jomar Bioscience | ant-pm2 | Combination of antibacterial and antifungal compounds to protect cell cultures from contaminations |
Prostaglandin E2 (PGE2) | Tocris | 2296 | support proliferation of cells |
Rotary tube mixer | Ratek | RSM7DC | |
R-spondin 1 conditioned media | In-house | WNT signalling regulator | |
SB202190 | Jomar Bioscience | s1077-25mg | Selective p38 MAP kinase inhibitor |
Scale | |||
Scalpel handle | Livingstone | WBLDHDL03 | |
Scalpels, #11 blade | Medical and Surgical Requisites | EU-211-1 | |
Serological pipettes (5, 10, 25 mL) | |||
Specimen Waste Bags | Medical Search | SU09125X16 | |
Urine specimen jar | |||
Y27632 | Jomar Bioscience | s1049-10mg | Selective ROCK inhibitor. Increases survival of dissociated epithelial cells |