Protocol for generation and utilization of patient-derived organoids from multimodal specimen

Summary

Patient-derived organoids (PDOs) are robust preclinical models for precision oncology. However, most existing protocols depend on surgically resected specimens, limiting their applicability to patients who are ineligible for surgery. Here, we present a protocol for generating PDOs from various clinically accessible specimens—including biopsies (endoscopic ultrasound-guided fine needle biopsy [EUS-FNB], percutaneous liver biopsy [PLB], ascites, and pleural effusion)—across cancer types. We describe steps for specimen transport, tumor cell isolation, culture, biobanking, and high-throughput drug screening, supporting reproducible PDO applications in translational research across diverse clinical settings.

For complete details on the use and execution of this protocol, please refer to Lee et al.¹

Graphical abstract

Highlights

• •Protocol for PDO establishment across cancer types
• •Standardized workflows for PDO generation, cryopreservation, and drug screening
• •Applicable to surgical and biopsy samples, including EUS-FNB, PLB, and body fluids
• •Guidelines to overcome practical challenges in organoid establishment

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Patient-derived organoids (PDOs) are robust preclinical models for precision oncology. However, most existing protocols depend on surgically resected specimens, limiting their applicability to patients who are ineligible for surgery. Here, we present a protocol for generating PDOs from various clinically accessible specimens—including biopsies (endoscopic ultrasound-guided fine needle biopsy [EUS-FNB], percutaneous liver biopsy [PLB], ascites, and pleural effusion)—across cancer types. We describe steps for specimen transport, tumor cell isolation, culture, biobanking, and high-throughput drug screening, supporting reproducible PDO applications in translational research across diverse clinical settings.

Before you begin

This protocol offers a robust and flexible method for establishing patient-derived organoids (PDOs) from a range of clinical specimens, including surgically resected tissues, core needle biopsies (e.g., EUS-FNB, PLB, punch), and liquid biopsies (e.g., malignant ascites or pleural effusions). Prior to specimen collection, ensure that ethical approval has been obtained and informed consent has been acquired from all patients.

Efficient tissue dissociation is a critical step for successful PDO establishment. In our protocol, we used Miltenyi’s Tumor Dissociation Enzyme Kit in combination with the gentleMACS Octo Dissociator with Heaters for surgical specimens. This setup enabled standardized mechanical and enzymatic dissociation under controlled temperature conditions, offering high reproducibility and throughput—particularly beneficial in biobanking settings dealing with numerous samples.^1,2,3,4

However, the use of gentleMACS is not mandatory. In laboratories without access to this instrument, a standard shaking incubator can be used for enzymatic digestion. Additionally, the Miltenyi enzyme kit may be substituted with more commonly available dissociation enzymes such as collagenase and/or dispase. For small specimens like biopsies or liquid samples, manual mechanical dissociation (e.g., gentle pipetting or tapping) combined with enzymatic incubation at 37°C is sufficient and often preferable.

Notably, this protocol accommodates the unique handling requirements of low-input specimens and is optimized for both solid and fluid tumor sources. The ability to generate organoids from liquid biopsies enables functional studies even in patients with advanced or inoperable diseases.⁵ Through practical optimizations derived from extensive experience with diverse clinical samples, this protocol provides a reproducible and accessible foundation for PDO establishment across a variety of research and clinical settings.

Note: This protocol involves the handling of human tumor-derived specimens. According to standard biosafety guidelines, all procedures should be conducted in a Class II biological safety cabinet to ensure the safety of laboratory personnel and the surrounding environment.

Institutional permissions

The specimens used in this protocol were collected at the National Cancer Center, Korea. The protocol was approved by the Institutional Review Board of the National Cancer Center of Korea (Approval No.: NCC2017-0122, NCC2020-0231, NCC2020-0259, NCC2020-0290, NCC2020-0337, NCC2021-0232), and written informed consent was obtained from all patients prior to specimen collection.

Reagent preparation for organoid culture

Timing: 10 min (for step 1)

Timing: 30 min (for step 2)

Timing: 30 min (for step 3)

Timing: 1 h for heat inactivation of FBS; 4–6 h for thawing heat-inactivated FBS at 4°C; 10 min for preparation of freezing medium (for step 4)

Timing: 30 min (for step 5)

Timing: 4–6 h for BME thawing to fully liquefy; 10 min for aliquot (for step 6)

• 1.
  Preparation of tissue transfer media.

  • a.
    Take 50 mL of serum-free RPMI 1640 medium.
  • b.
    Add 100 μL of 500× primocin (50 mg/mL).
  • c.
    Mix the solution thoroughly.
  • d.
    Aliquot 4 mL of the mixture into each 5.0-mL Eppendorf Tube.
  • e.
    Store the aliquots at 4°C for up to 2 weeks.
• 2.
  Preparation of tumor tissue dissociation enzymes for cell isolation.

  Note: Follow the manufacturer's guidelines for the preparation and storage of the Human Tumor Dissociation Kit (Miltenyi, Cat#130-095-929). Ensure that all enzyme aliquots avoid freeze-thaw cycles. All aliquots should be stored in light-protective brown tubes.

  The user manual is found at the following link: https://static.miltenyibiotec.com/asset/150655405641/document_qlvs47ia0p6lv0mm12nmpm0i46?content-disposition=inline.

  • a.
    Reconstitute Enzyme H.

    • i.
      Add 3 mL of serum-free RPMI 1640 to each of the two vials containing lyophilized Enzyme H.
    • ii.
      Mix each vial thoroughly by gentle pipetting or inversion.
    • iii.
      Filter each solution through a 0.2 μm syringe filter.
    • iv.
      Aliquot 200 μL per tube.
  • b.
    Reconstitute Enzyme R.

    • i.
      Add 2.7 mL of serum-free RPMI 1640 to the vial containing lyophilized Enzyme R.
    • ii.
      Mix thoroughly and aliquot 100 μL per tube.
  • c.
    Reconstitute Enzyme A.

    • i.
      Add 1 mL of Buffer A (provided in the kit) to the vial containing lyophilized Enzyme A.
    • ii.
      Aliquot 30 μL per tube.
  • d.
    Store all enzyme aliquots at −20°C. Reconstituted enzymes are stable for up to 6 months.
• 3.
  Prepare the organoid growth medium.

  Note: Organoid growth medium should be customized for each cancer type by incorporating specific growth factors that promote organoid generation and proliferation, reflecting the unique niche factors of each type. This protocol provides the optimized medium composition for six types of cancer, each successfully established and expanded as organoids, based on previously published studies.^6,7,8,9 Additionally, the protocol can be adapted for other cancer types by adjusting the growth medium composition as needed (Problem 2).

  • a.
    Thaw all stock solutions of the components stored at −20°C.

    Note: Stock solutions are prepared at the indicated concentrations, sterile-filtered, aliquoted, and stored at −20°C.

  • b.
    Prepare 500 mL of the human tumor organoid basic medium following the composition listed in the table in materials and equipment.
  • c.
    Aliquot the organoid basic medium into 50 mL portions.
  • d.
    Prepare the final organoid growth medium by adding tumor-type-specific supplements to the basic medium, as described in the ‘materials and equipment’ section.
• 4.
  Prepare 50 mL of the organoid freezing medium.

  • a.
    Heat-inactivation of FBS.

    • i.
      Incubate at 60°C for 30 minutes in a water bath to heat-inactivate the FBS.

      Note: Gently mix every 5–10 minutes during incubation to ensure even heating.

    • ii.
      Cool to 20°C–25°C and filter the heat-inactivated FBS using a 0.2 μm bottle-top filter under sterile conditions.
    • iii.
      Aliquot the FBS into 45 mL sterile conical tubes and store at −20°C.
  • b.
    Take the 45 mL aliquot of heat-inactivated FBS stored at −20°C and thaw slowly at 4°C.
  • c.
    Using a serological pipette attached to a pipette aid, add 5 mL of DMSO.
  • d.
    Add 500 μL of 100× Y-27632, mix thoroughly, and store at 4°C for up to 1 month.
• 5.
  Prepare 500 mL of 0.1% BSA-PBS solution.

  • a.
    Add 0.5 g of BSA to 500 mL of sterile 1× PBS prepared by autoclaving.
  • b.
    Mix the solution until it becomes completely homogeneous.
  • c.
    Filter the solution using a 0.2-μm-pore bottle top vacuum filter system.
  • d.
    Aliquot 50 mL of each solution into 50 mL conical tubes and store at 4°C for up to 1 month.
• 6.
  Aliquot the basement membrane extract (BME).

  • a.
    Thaw 5 mL of BME gradually at 0–4°C until completely liquefied.
  • b.
    Aliquot 1 mL of the thawed BME into 1.7-mL tubes.
  • c.
    Store the required volume at 4°C for immediate use and return the remaining aliquots to −20°C for future use.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Biological samples
Specimens derived from patients with cancer	N/A	N/A
Chemicals, peptides, and recombinant proteins
RPMI 1640	HyClone	SH30027.01
MACS tissue storage Solution	Miltenyi	130-110-008
Tumor dissociation kit, human	Miltenyi	130-095-929
Advanced DMEM/F-12	Gibco	12634-010
B-27 supplement (50×), serum free	Invitrogen	17504-001
GlutaMAX supplement	Gibco	35050-061
HEPES (1 M)	Gibco	15630-080
Noggin-Fc fusion protein conditioned medium	U-Protein Express BV	N002-200ML
Nicotinamide	Sigma	N0636-100G
N-2 supplement (100×)	Invitrogen	17502-048
N-acetyl-L-cysteine	Sigma	A9165
Recombinant human R-spondin 1 protein	Qkine	Qk006
WNT Surrogate-Fc fusion protein	U-Protein Express BV	N001-500UG
A 83-01	Tocris	2939
Recombinant human FGF-10	PeproTech	100-26
Animal-free recombinant human EGF	PeproTech	AF-100-15
Primocin	InvivoGen	ant-pm-2
SB 202190	Tocris	1264
Forskolin	Tocris	1099
Y-27632 dihydrochloride	Tocris	1254
Gastrin 1 human	Sigma	G9020
Albumin, ultra-pure bovine serum	GenDEPOT	A0100-010
10x Dulbecco’s phosphate-buffered salines (D-PBS)	Welgene	LB201-02
DMSO	Sigma	D2650-100ML
Fetal bovine serum	Gibco	16000044
Cultrex reduced growth factor basement membrane extract, type 2, PathClear	R&D Systems	3533-005-02
DNase I, RNase-free (1 U/μL)	Thermo Fisher Scientific	EN0521
Ficoll-Paque PREMIUM	Cytiva	17544202
LIVE/DEAD viability/cytotoxicity kit, for mammalian cells	Invitrogen	L3224
TrypLE Express enzyme (1X), phenol red	Gibco	12605028rbc
RBC lysis buffer (10X)	BioLegend	420301
EVE trypan blue stain 0.4%	NanoEntek	EBT-001
e-Myco mycoplasma PCR detection kit	Intron	25235
Critical commercial assays
CellTiter-Glo 3D cell viability assay	Promega	G9683
Software and algorithms
GraphPad Prism v.8.4	GraphPad	https://www.graphpad.com/scientific-software/prism/
Other
PIPETMAN L 4-pipette kit	Gilson	F167370
Eppendorf Tubes 5.0 mL	Eppendorf	0030119401
Eppendorf Safe-Lock Tubes, light-protective brown	Eppendorf	0030121155
5QT performance cooler box	Coleman	4992826454876
Pipette aid	Falcon	FA.7590
15 mL conical tube	Falcon	FA.352096
50 mL conical tube	Falcon	FA.352070
24-well plates	Corning	353047
1.7 mL tube	Axygen	MCT-175-C
10 μL pipette tips	Axygen	AX.T-300
200 μL pipette tips	Axygen	AX.T-200-Y
1000 μL pipette tips	Axygen	AX.T-1000-B
5 mL serological pipette	Falcon	357543
10 mL serological pipette	Falcon	356551
25 mL serological pipette	Falcon	357535
gentleMACS C tubes	Miltenyi	130-096-334
Forceps	Sewoon Medical	4401-002
Surgical blade#21	Feather	HFE-SB21
Handle for surgical blade	Surgicrafts	HSC-0610400
Iris scissors, straight, 10 cm	Kasco	5-005
Pasteur pipettes, 230 mm	Hilgenberg	HG.3150102
EVE slide	NanoEntek	EVS-050
Millex PVDF syringe filter	Millipore	SLGV033N
Syringe, 10 mL, sterile	N/A	N/A
Ice bucket	N/A	N/A
Protein LoBind tubes	Eppendorf	0030108116
Reservoir, 25 mL	SPL	21102
Reservoir, 50 mL	SPL	23050
96-well clear round bottom ultra-low attachment microplate	Corning	7007
384-well flat clear bottom white polystyrene TC-treated microplates	Corning	3765
CoolCell FTS30 cell freezing container	Corning	CLS432009
Spherical Dewar flask	KGW	KG.1211
Floating tube rack	LK Lab Korea	R01-76-241
Nalgene Rapid-Flow sterile disposable filter units with PES, 0.2 μm	Thermo Fisher Scientific	566-0020
Nunc biobanking and cell culture cryogenic tube	Thermo Fisher Scientific	368632
Polypropylene cryobox 10 × 10 grid	Crystal	PPP2-100
gentleMACS Octo dissociator with heaters	Miltenyi	130-096-427
Biological safety cabinets	N/A	N/A
CO2 incubator	Thermo Fisher Scientific	3111
37°C water bath	Thermo Fisher Scientific	N/A
Centrifuge 5810R	Eppendorf	5811000015
Micro centrifuge 5415D	Eppendorf	5415D-KR-R
Inverted microscope	Olympus	N/A
Inverted fluorescence microscope	Zeiss	ZEISS Axio Observer
Automatic Cell Counter_EVE PLUS	NanoEntek	EVE-MC2
Vacuum pump	N/A	N/A
Picus electronic pipette, 8 channel	Sartorius	735361
Operetta CLS high-content analysis system	PerkinElmer	Operetta CLS
Spark multimode microplate reader	Tecan	SPARK 10M

Materials and equipment

Human tumor organoid basic medium

Reagent	Stock concentration	Final concentration	Amount
Advanced DMEM/F-12	N/A	N/A	470 mL
50× B-27 supplement	50×	1×	10 mL
100× GlutaMAX	100×	1×	5 mL
100× HEPES	1 M	10 mM	5 mL
100× Noggin-Fc Fusion Protein Conditioned Medium	100×	1×	5 mL
100× Nicotinamide	1 M	10 mM	5 mL
80× N-acetyl cysteine, NAC	100 mM	1.25 mM	6.25 mL
10000× R-spondin1	500 μg/mL	50 ng/mL	50 μL
1000× WNT Surrogate-Fc Fusion Protein	10 μg/mL	100 ng/mL	500 μL
5000× A83-01	2.5 mM	500 nM	100 μL
1000× Recombinant Human FGF10	100 μg/mL	100 ng/mL	500 μL
2000× Recombinant Human EGF	100 μg /mL	50 ng/mL	250 μL
500× Primocin	50 mg/mL	100 μg/mL	1 mL
Total	N/A	N/A	500 mL

Organoid basic medium can be stored at 4°C up to 1 month.

Note: Add 100× Y-27632 (5 mM) to the organoid culture medium at a final concentration of 1× (50 μM) for the first 3 days after thawing frozen organoids.¹⁰

Human oral cancer organoid growth medium

Reagent	Stock concentration	Final concentration	Amount
Basic medium	N/A	N/A	50 mL
500× Forskolin	5 mM	10 μM	100 μL
Total	N/A	N/A	50 mL

Oral cancer organoid growth medium can be stored at 4°C up to 1 month.

Human pancreatic cancer organoid growth medium

Reagent	Stock concentration	Final concentration	Amount
Basic medium	N/A	N/A	50 mL
1000× Gastrin-1	50 μM	50 nM	50 μL
Total	N/A	N/A	50 mL

Pancreatic cancer organoid growth medium can be stored at 4°C up to 1 month.

Human colon cancer organoid growth medium

Reagent	Stock concentration	Final concentration	Amount
Basic medium	N/A	N/A	50 mL
5000× Gastrin-1	50 μM	10 nM	10 μL
333× SB 202190	1 mM	3 μM	150 μL
100× N2 supplement	100×	1×	500 μL
Total	N/A	N/A	50 mL

Colon cancer organoid growth medium can be stored at 4°C up to 1 month.

Human gastric cancer organoid growth medium

Reagent	Stock concentration	Final concentration	Amount
Basic medium	N/A	N/A	50 mL
5000× Gastrin-1	5 μM	1 nM	10 μL
Total	N/A	N/A	50 mL

Gastric cancer organoid growth medium can be stored at 4°C up to 1 month.

Human cholangiocarcinoma and gall bladder cancer organoid growth medium

Reagent	Stock concentration	Final concentration	Amount
Basic medium	N/A	N/A	50 mL
1000× Gastrin-1	50 μM	50 nM	50 μL
500× Forskolin	5 mM	10 μM	100 μL
100× N2 supplement	100×	1×	500 μL
Total	N/A	N/A	50 mL

Cholangiocarcinoma and gall bladder cancer organoid growth medium can be stored at 4°C up to 1 month.

Step-by-step method details

Establishment of organoids from surgical tissue

Timing: 20 min for sample processing; 1 h 30 min for tissue dissociation and organoid seeding; 30 min for 3D droplet solidification

This step outlines the process of collecting the patient-derived tumor tissue, isolating tumor cells, and establishing organoids (Figure 1A). The tumor tissue is cut into several cubes, sized approximately 0.5 × 0.5 cm. One is used for organoid generation, and the others are stored for genomic, proteomic, and further analyses.

CRITICAL: After tumor tissue is obtained, any external contamination must be prevented. Moreover, the proportion of tumor cells and the composition of the tumor microenvironment are key to determining the success of organoid establishment. High immune or stromal content and low tumor cellularity significantly reduce success rates.

• 1.
  Collection and initial processing of resected tumor tissue.

  Note: If the tissue specimen can be processed within 2–4 h of receipt, use transfer media for its collection. For storage periods exceeding 4 h, use Miltenyi’s MACS Tissue Storage Solution to collect and store the specimen.

  • a.
    Prepare aliquoted transfer media, sterilized forceps, surgical blades, and ice in a cooler box for specimen transport.
  • b.
    Immediately place the surgically resected tissue into the 5-mL Eppendorf Tube containing the transfer media and keep it on ice.
  • c.
    Place the tissue specimen on a petri dish using sterile forceps and assess its properties and stiffness.
  • d.
    Remove non-tumor components (e.g., adipose, epithelial, or muscle tissue), leaving only the tumor-containing area (Figure 2A).
  • e.
    Cut the tumor portion into 0.5 × 0.5 cm cubes (Figure 2B).
• 2.
  Dissociation of the tissue for tumor cell isolation.

  Note: This protocol utilizes the automated tissue dissociation method using the GentleMACS Octo Dissociator device from Miltenyi.

  • a.
    Take one vial each of tumor dissociation enzymes H, R, and A stored at −20°C and thaw them at 4°C.
  • b.
    Place one tissue on a petri dish preloaded with 100 μL of serum-free RPMI 1640 and mince the tissue using a surgical blade (Feather, Cat#HFE-SB21) until the texture becomes soupy (Figure 2C).

    CRITICAL: Proper mincing increases the surface area for enzyme access and improves cell yield. Continue mincing until the texture becomes soupy.

  • c.
    Open the lid of the GentleMACS C Tube and carefully tilt the Petri dish containing minced tissue against the tube opening.
  • d.
    Using a 5-ml serological pipette, dispense 4.5 mL of serum-free RPMI media with sufficient force to gently push the tissue into the GentleMACS C Tubes.

    CRITICAL: Do not pipette the tissue up and down, as this can cause it to stick to the pipette and reduce recovery. Use the medium flow to gently push the tissue into the tube.

  • e.
    Add 200 μL of Enzyme H, 100 μL of Enzyme R, and 25 μL of Enzyme A into the C tube.
  • f.
    Tightly close the screw cap and mount the C tube vertically onto the GentleMACS Octo Dissociator with the cap facing downward. The heater should then be placed over it.
  • g.
    Run the “37C_h_TDK_3” program.

    Note: The program of GentleMACS Octo Dissociator should be selected based on tissue stiffness observed in Step 1-c, as well as the characteristics specific to each tumor type (Table 1).

    Note: The attached heater maintains incubation at 37°C, while physical dissociation is facilitated by the C tube rotor.

  • h.
    After 40 min, abort the program and detach the C tube from the device.

    CRITICAL: If under-dissociated tissues are observed in the C tube, incubate the specimen for an additional 10 min. However, the total incubation time should not exceed 1 h. Extended incubation does not enhance the dissociation of fibrous or connective tissue and may damage tumor cells. As such, the decision for further incubation must be made with due discretion based on the physical properties of the tissue (Problem 1).

  • i.
    Transfer the cell suspension from the C tube to a 15-mL tube.

    CRITICAL: When transferring a cell suspension using a serological pipette, cells or tissue may adhere to its surface, leading to loss. Therefore, the suspension should be poured directly into the tube.

    CRITICAL: To increase the success rate of tumor organoid establishment, skip using a strainer to prevent the loss of tumor cells trapped between tissue fragments, as shown in Figure 3.

  • j.
    Add 5 mL of serum-free RPMI 1640 to the original C tube, close the lid, and shake to collect any remaining cells.
  • k.
    Open the C tube and pour its contents into the 15-mL tube containing the initial suspension.

    Note: Conduct the procedures as quickly as possible. FBS-containing media are conventionally used to inactivate dissociation enzymes; however, the aim of this protocol is to produce organoids in an FBS-free environment.

  • l.
    Centrifuge the cell suspension at 360 × g for 5 min at RT.
  • m.
    Aspirate the supernatant and add 5 mL of PBS for cell pellet washing using a 5-ml serological pipette.
  • n.
    Centrifuge again at 360 × g for 5 min at 25°C and aspirate the supernatant.

    Optional: If red blood cells (RBCs) visibly occupy around 50% of the total volume of the cell pellet, proceed with RBC lysis.

    • i.
      Add 1 mL of 1× RBC lysis buffer to the 15-mL tube containing the pellet.
    • ii.
      Gently resuspend the pellet by tapping, then incubate at RT for 1 min.

      Note: Initially, the solution will appear red and cloudy; it will turn red and clear after incubation.

      CRITICAL: If many RBCs are present, incubate for additional time. However, be cautious, as prolonged incubation can cause damage to tumor cells. The recommended time is not more than 10 min.

    • iii.
      Add 9 mL of PBS using a 10-mL serological pipette to dilute the RBC lysis buffer.
    • iv.
      Centrifuge at 360 × g for 5 min at 25°C.
  • o.
    Resuspend the cell pellet with 1 mL of PBS using a 1000-μL pipette tip and transfer the cell suspension to a 1.7-mL Eppendorf Tube.
• 3.
  Counting the cells and seeding for 3D culture.

  • a.
    Mix 10 μL of cell suspension with 10 μL of trypan blue solution.
  • b.
    Load 10 μL of the mixture onto a counting slide compatible with an automated cell counter.
  • c.
    Perform the cell counting.
  • d.
    Determine the number of seeding droplets and calculate the required volume of BME.

    Note: When single-cell dissociation is achieved, seed 1 × 10⁵ cells in 40 μL BME per dome in a 24-well plate. However, since complete dissociation is often not achieved, the pellet volume may be larger than the counted cell number. Then, estimate the number of domes to seed based on the pellet volume. For example, a tumor tissue of approximately 0.5 × 0.5 cm typically yields material for seeding 4 wells of 24-well plate. This is not a fixed rule and should be adjusted based on tissue composition and dissociation efficiency.

    Optional: If enough cells have been obtained to seed six or more domes, the remaining cells can be cryopreserved in freezing media for future use.

  • e.
    Place BME aliquots on ice.
  • f.
    Centrifuge the cell suspension in a 1.7-mL Eppendorf Tube at 360 × g for 5 min at RT.
  • g.
    Remove the supernatant completely using a 1000-μL pipette tip.
  • h.
    Slowly and carefully resuspend the cell pellet with the calculated volume of BME (problem 6).
  • i.
    Place a 40 μL droplet of BME-cell suspension in the center of each well of the 24-well plate. (problem 6 and 7).

    Note: If partially undissociated tissues are present, cutting the end of the pipette tip with sterilized scissors can facilitate droplet seeding.

    CRITICAL: Bubbles trapped in the 3D dome can compromise structural integrity and lead to dome collapse during incubation after media is added. To prevent this, pipette BME slowly and carefully, as it is highly viscous. Although small air bubbles are generally not problematic, if large (macro) bubbles are formed, they should be removed immediately after dome formation using a 10 μL or 200 μL pipette tip connected to a pipette to aspirate the air. This step must be performed before the dome solidifies.

  • j.
    Observe the overall landscape of the seeded 3D dome under the microscope (Figure 4).
  • k.
    Incubate the 24-well plate at 37°C and 5% CO₂ for 30 min to solidify.
  • l.
    Add 500 μL of pre-warmed organoid growth medium per well.

    Note: To avoid disrupting the solidified BME dome, gently add the medium by dispensing it along the side wall of the well.

  • m.
    Incubate the plate at 37°C and 5% CO₂.

Figure 1.

Schematic overview of the detailed steps of patient specimen processing for establishing PDOs

(A) Collected surgically resected tissue specimens were processed by mincing, followed by enzymatic and mechanical dissociation.

(B) Collected biopsy specimens were minced with sterile scissors, followed by enzymatic dissociation and incubation.

(C) Liquid biopsy specimens were processed by density gradient centrifugation to isolate mononuclear cells.

Figure 2.

Procedure for processing resected tissue specimens to establish PDOs

(A) After collecting the tissue, any region presumed to be normal are removed to isolate tumor cells for establishing tumor PDOs. The scale bar represents 1 cm.

(B) The tumor tissue is cut into fragments larger than 0.5 × 0.5 cm, with one fragment used for organoid generation and the remaining pieces stored according to their intended use.

(C) To increase the tissue surface area for effective enzyme dissociation, the tissue is minced thoroughly with a surgical blade until it reaches a slurry-like consistency. The scale bar represents 1 cm.

Table 1.

Tumor type-specific guidelines for tissue dissociation

Tumor type	Tissue stiffness	Dissociation program	Dissociation time
Oral cavity	Tough	37C_h_TDK_3	30 min
Pancreas	Tough	37C_h_TDK_3	40 min
Colon	Soft	37C_h_TDK_1	40 min
Gastric	Moderate	37C_h_TDK_1	40 min
Biliary tract	Tough	37C_h_TDK_3	40 min
Gallbladder	Moderate-tough	37C_h_TDK_3	30 min

Figure 3.

The landscape of PDOs generated within stromal tissue

(A) Microscopic image of fibrous components and dissociated tumor cells after enzymatic and mechanical dissociation.

(B) Example of tumor organoids growing within stromal tissue. Omitting the strainer step allows organoids to form from tumor cells embedded within stromal tissue, thereby minimizing cell loss from limited patient samples. The white arrow indicates the organoid; the black arrow indicates the stromal tissue. The scale bar represents 100 μm.

Figure 4.

Microscopic visualization of tumor and stromal cell distribution within BME

This figure illustrates various possible compositions that can be observed microscopically: (A): Appearance of single tumor cells; (B): Aggregated tumor cells resembling organoids since their separation from the tissue; (C): Tumor cells observed between tissue fragments; (D): Tissue fragments anticipated to cease growth due to differentiation; (E): Small-sized cells, presumed to be immune cells; (F): Air bubbles may form during embedding.

Organoid establishment from biopsy tissue (endoscopic ultrasound-guided fine needle biopsy, percutaneous liver biopsy, and punch biopsy)

Timing: 1 h for organoid seeding; 30 min for 3D droplet solidification

This step describes the process for generating organoids from EUS-FNB, PLB, and punch biopsy specimens, primarily collected from metastatic or locally advanced tissues where surgery is not feasible (Figure 1B).

Note: EUS-FNB specimens generally tend to contain smaller and softer tissue fragments, often with a higher presence of RBCs. Conversely, PLB and punch biopsy specimens are typically obtained using a relatively large needle through the skin, resulting in larger tissue pieces with higher density. The protocol should be adjusted to these tissue characteristics accordingly.

• 4.
  Collection of biopsy tissue.

  • a.
    Prepare the transfer media aliquot and ice in a cooler box for specimen transport.
  • b.
    Immediately place the biopsy specimen into a 5-mL Eppendorf Tube containing the transfer media and place on ice.
• 5.
  Dissociation of biopsy tissue for tumor cell isolation.

  • a.
    Wash the tissue.

    • i.
      Place the tube on a rack and wait for 5 min to allow the tissue to sink by gravity.
    • ii.
      When the tissue settles to the bottom, remove the supernatant using a 1000-μL pipette.
    • iii.
      Add 2 mL of fresh transfer media containing 100 μg/mL of primocin.
    • iv.
      Repeat i–iii twice.

      Note: If the tissue has been dispersed and does not maintain its shape, collect it by centrifuging at 360 × g for 5 min.

  • b.
    Transfer the specimen to a 1.7-mL Protein LoBind Tube containing 1 mL of serum-free RPMI using sterilized forceps.
  • c.
    Chop the tissue inside the tube using sterilized scissors (Kasco, Cat#5-005).
  • d.
    Thaw one vial of each tumor dissociation enzymes on ice.
  • e.
    Add 40 μL of Enzyme H, 20 μL of Enzyme R, and 5 μL of Enzyme A to the tube.

    Optional: If the tissue is highly viscous after chopping, add 1 μL of DNase I to reduce gDNA content.

  • f.
    Incubate the tubes at 37°C with rotation for 15 min.

    Note: If a rotator with temperature control is not available, incubate statically at 37°C and tap every 5 min.

  • g.
    Centrifuge again at 360 × g for 5 min at RT and remove the supernatant.

    Optional: If RBCs visibly occupy around 50% of the total volume of the cell pellet, proceed with the RBC lysis.

    • i.
      Add 1 mL of 1× RBC lysis buffer to the 1.7-mL Eppendorf Tube containing the pellet.
    • ii.
      Gently resuspend the pellet by tapping, then incubate at RT for 1 min.

      Note: Adjust the incubation time to ensure the solution becomes transparent, as seen in Step 2-n-ii.

    • iii.
      Centrifuge at 7,500 × g for 1 min at 25°C.
    • iv.
      Remove the supernatant and resuspend the pellet with 1 mL of PBS.
    • v.
      Repeat steps iii and iv.
• 6.
  Counting the cells and seeding for 3D culture.

  • a.
    Return to Step 3 and proceed as described.

Organoid establishment from liquid biopsy specimens (ascites/pleural effusion)

Timing: 1 h 30 min for organoid seeding; 30 min for 3D droplet solidification

This Section describes the protocol for establishing organoids using tumor cells from ascites and pleural effusion specimens.⁵ The key aspect of this process is the use of density gradient centrifugation to isolate the mononuclear layer containing tumor cells (Figures 1C and 5).

• 7.
  Collecting the liquid biopsy specimen.

  • a.
    Collect specimens of ascites or pleural effusions from patients through paracentesis via percutaneous drainage (Figure 5A).
  • b.
    Store the collected specimens (50–100 mL) at 4°C under sterile conditions until the procedure.
• 8.
  Isolating the tumor cells from the specimens.

  • a.
    Transfer the collected specimens into a 50-mL tube (Figure 5B).
  • b.
    Centrifuge at 400 × g for 10 min at 25°C.
  • c.
    Dispense 7 mL of Ficoll-Paque solution into a new 15-ml tube.
  • d.
    After centrifugation, check the amount of settled pellet and aspirate the supernatant.
  • e.
    Gently tap the cell pellet and add 7 mL of serum-free RPMI to fully resuspend.
  • f.
    Layer the cell resuspension dropwise onto the Ficoll-Paque solution in the 15-ml tube (Step 8-c) using a serological pipette (Figure 5C).

    CRITICAL: Proceed slowly to avoid mixing the layers.

  • g.
    Set centrifuge brake to level 5, and centrifuge at 400 × g for 30 min at 25°C.

    CRITICAL: Rapid deceleration can disturb layer separation; slow braking is essential.

    Note: After centrifugation, four layers should appear. The second layer (Figure 5D) contains mononuclear cells including tumor cells.

  • h.
    Aspirate the top media layer, leaving ∼1 mL.
  • i.
    Attach a Pasteur pipette to the pipette aid and use it to carefully transfer the mononuclear layer to a 50 ml tube (Figure 5E).

    Note: It is acceptable if a small amount of solution from other layers is also aspirated. The priority is to collect as many cells as possible.

  • j.
    Wash the cells by adding serum-free RPMI 1640 medium to the 50 mL-tube, using at least three times the volume of the collected cell suspension (Figure 5F).
• 9.
  Counting the cells and seeding for 3D culture.

  • a.
    Proceed as described in steps 2-l through 3 (Figures 5G and 5H).

Note: Liquid biopsy samples often contain a higher proportion of immune cells compared to solid tissue specimens, and if RBC lysis is not performed, the total cell count may be overestimated. Therefore, when counting cells, set the threshold on the automated counter to include only cells larger than 9 μm, and use this adjusted count for determining the number of droplets to seed.

Figure 5.

Isolation process of tumor cells from liquid biopsy specimens

(A and B) The collected liquid biopsy specimen is poured into a 50-mL conical tube and centrifuged to obtain a pellet.

(C) The pellet is then resuspended in serum-free RPMI 1640 and carefully dispensed over the Ficoll without mixing.

(D and E) The mononuclear cell layer isolated by centrifugation is aspirated using a Pasteur pipette attached to a pipette aid to minimize cell loss. If a high proportion of RBCs are present, an additional lysis step can be performed.

(F–H) The cell pellet should be washed twice and then mixed with BME to prepare for seeding in droplets.

Monitoring organoid formation and assessing morphology

Timing: 20 min

This process outlines the methods for assessing the success or failure of organoid establishment and determining passage time based on the observed morphological characteristics.

• 10.
  Organoid observation.

  • a.
    One day after seeding, observe the organoids every 1–2 days using a microscope.

    Note: Beginning the day after seeding, the following should be carefully monitored: presence of organoid formation (Problem 3), contamination, cell distribution presence of stromal components (Problem 4), and BME dome stability (e.g., collapse, breakage, detachment from the plate) (Problem 5). Record daily observations and notable features on a log sheet to support troubleshooting if establishment fails. To monitor growth, capture images of organoids under a ZEISS microscope at 10×, 20×, and 40× magnification in bright-field mode at least twice a week after seeding.

  • b.
    Select organoids that have reached the appropriate state for passaging.

    Note: The time required for one passage may differ between early (p.1–4) and late organoids (>p.5) after stabilization.

    • i.
      Monitor the growth rate and density of the organoids.

      Note: Low density or lack of growth in early passages may indicate failure (Figures 6A–6E) (problem 2, 3, and 4). The formation and growth of organoids should be observed, indicating successful establishment (Figures 6F–6J).

      Note: In early passages, observation for up to 1 month may be needed to determine the success or failure of establishment (problem 2).

      CRITICAL: If the BME droplet has collapsed due to the presence of a large amount of tissue, perform passaging immediately following the procedure described in Step 12, and monitor organoid formation.

    • ii.
      Stabilized organoid lines can be cultured for 7–14 days before passaging (Figure 7).

      Note: The criteria for determining the passage timing are based on organoid diameter, estimated using the scale bar from ZEISS microscope images, and visual density, defined as approximately 50% coverage of the field of view, as shown in Figure 7. These parameters may vary depending on organoid morphology, as described in Table 2.

Figure 6.

Assessment of growth failure or success in newly established PDO attempts

(A–E) The primary causes of failure in establishing tumor organoids from patient-derived samples are typically low tumor cellularity or an environment unsuitable for tumor cell growth. This can be categorized into five representative cases: (A) No tumor cell growth (problem 2)or low cellularity of tumor cells (problem 3); (B) Observation of differentiated tissue fragments that do not dissociate further (problem 1 and 3); (C) Significantly higher proportion of immune cells compared to the number of tumor cells (problem 3); (D) Overgrowth of fibroblasts (problem 4); (E) Excessive cell count embedded in BME (problem 5).

(F–J) Representative images illustrating the successful establishment of tumor organoids. (F, G) Growing established tumor organoids; (H) Organoids generated within stromal tissues; (I) Organoids transformed from tissue fragments; (J) Immune cells and established organoids. Black arrowheads indicate organoids; white arrowheads indicate stromal tissue. Solid black arrows denote fibroblasts. Hollow arrows indicate immune cells.

The scale bar represents 100 μm.

Figure 7.

Representative images of PDOs suitable for passaging

Determining the passage time is crucial for maintaining the characteristics of organoids and amplification for long-term cultures. Passage duration varies based on the tumor type and the morphological characteristics of the organoids, considering factors such as the diameter and density of the organoids. The scale bars represent 100 µm in the upper panels and 50 µm in the lower panels.

Table 2.

Criteria for organoid passage

Morphology	Size (maximum)	Period
Dense, Grape-like	∼150 μm	Average: 2 weeks Maximum: 4 weeks
Cystic, Balloon-like	∼300 μm	Average: 2 weeks Maximum: 4 weeks

Organoid maintenance and passaging

Timing: 10 min for medium change; 1 h 30 min for passaging; 30 min for 3D droplet solidification

This process outlines the procedures for dissociating organoids into single cells, reseeding them for continued cultures, and banking amplified cells for long-term storage (Figure 8). This section focuses on ensuring that researchers can maintain organoid cultures over time and access the cells as needed for future experiments.

Note: Regarding the maintenance of organoid cultures, short tandem repeat (STR) analysis should be performed when reaching passages 5, 10, and 20. This analysis is used to assess whether the STR profiles from the DNA extracted from the organoids match those from the gDNA extracted from blood or tumor tissue.¹¹ This process is used to evaluate the genetic consistency of the organoids during culture and verify their genetic similarity to the original tissue, acting as a genetic fingerprint. Simultaneously, Mycoplasma contamination is checked using a Mycoplasma PCR detection kit.¹²

• 11.
  Changing the organoid growth medium.

  • a.
    Aspirate the existing culture medium from each well and replace it with 500 μL of fresh organoid growth medium every 3 days.

Note: Aspirate the medium gently from the corner of the well to avoid drawing up the BME/organoids droplet.

• 12.
  Passaging of organoids.

  • a.
    Aspirate the existing culture medium from all wells of the same organoid line.
  • b.
    Harvest the organoids by adding 1 mL of cold 0.1% BSA-PBS to each BME dome and gently pipetting up and down 3–5 times using a 1000-μL pipette tip.

    Note: Rinsing pipette tips with 0.1% BSA-PBS helps prevent organoid adhesion and minimize sample loss.

  • c.
    Reuse the same 1 mL of solution to sequentially process up to six wells of the same organoid line.
  • d.
    After collecting domes, transfer the suspension into a 1.7-mL Eppendorf Tube.

    Note: If more than six wells are available, divide the total number of wells in half and use 1 mL of solution for each group.

  • e.
    Centrifuge at 7,500 × g for 1 min at 25°C using a microcentrifuge.

    Note: To ensure that organoids pass through the BME layer and settle at the bottom of the tube, conduct centrifugation at a high RCF for a brief period to minimize potential cell damage.

  • f.
    Once the solution separates into three layers, carefully aspirate the upper PBS layer, leaving 1–2 mm of the middle BME layer.

    Note: Following centrifugation, the solution typically separates into three layers due to differences in density: PBS at the top, BME in the middle, and the organoid cell pellet at the bottom.

    CRITICAL: If many organoids are trapped in the BME layer, leave more of the BME layer behind.

  • g.
    Add 600 μL of TrypLE into the tubes.
  • h.
    Scrape the bottom of tubes 10-15 times against the plastic surface to help loosen the pellet (Figure 8B).
  • i.
    Incubate the tube for 2 min at 37°C in a heat block.
  • j.
    Check the state of organoids dissociation under the microscope.
  • k.
    Repeat steps h-j until the organoids are dissociated into single cells, for up to a maximum of 15 minutes (Figures 8C–8E).

    Note: While physical dissociation can be performed by pipetting or tapping, we found that firmly scraping the tube against a plastic surface (e.g., sample box compartments) was the most effective method for rapid dissociation. This approach minimized TrypLE exposure time and reduced damage to organoids by enabling quicker and more complete dissociation, regardless of organoid morphology.

  • l.
    Remove the tube from the heat block and add 600 μL of serum-free Advanced DMEM/F12.
  • m.
    Centrifuge at 360 × g for 5 min at RT (problem 8).
  • n.
    Aspirate the supernatant, while ensuring not to disturb the organoid pellet.

    Note: To prevent accidental loss of the organoid pellet during aspiration, leave approximately 100 μL of supernatant at the bottom of the tube.

  • o.
    Add 1 mL of 0.1% BSA-PBS solution and resuspend the pellet.
  • p.
    Mix 10 μL of cell suspension with 10 μL of trypan blue solution.
  • q.
    Load 10 μL of the mixture onto a counting slide and perform the cell counting.
  • r.
    Determine the number of cells required for seeding.

    CRITICAL: Organoid density is essential in growth optimization.¹³ An excessive number of organoids can deplete culture resources, whereas a small number of organoids may not provide enough paracrine signals to support proper growth. Seeding is performed depending on tumor type and organoid characteristics. Well-growing organoids are seeded with 3 × 10⁴ cells per droplet using 40 μL of BME while slower-growing organoids are seeded with 5 × 10⁴ cells per droplet.

    Note: If the total cell count exceeds 5 × 10⁵ cells, seed four to six droplets and freeze the remaining cells for storage (Section 13).

  • s.
    Transfer the required volume of cell suspension to a new 1.7 mL Eppendorf Tube.

    Note: Using the total cell count in the 1 mL suspension, determine the volume that contains the required number of cells.

  • t.
    Continue as indicated in steps 3-e to 3-m.

Figure 8.

Practical steps for organoid passaging

Organoids are selected for passage time based on observation through microscopy (A). Harvested organoids undergo physical dissociation through scraping and chemical dissociation in a 37°C incubator (B), gradually leading to the dissociation into single cells (C–E). By counting the number of cells, the appropriate number of BME droplets is determined for seeding to grow organoids for the next passage (F).

Organoid cryopreservation and thawing

Timing: 30 min for cryopreservation; 1 h for thawing; 7 days for organoid culture; 1 h for organoid cell viability assay

• 13.
  Freezing organoids for long-term storage.

  • a.
    Label the cryovial with the organoid name, passage number, cell count, and date of freezing.
  • b.
    Centrifuge the tubes containing the remaining cells from step 12-s at 360 × g for 5 min at 25°C.
  • c.
    Remove the supernatant.
  • d.
    Add 300 μL of organoid freezing medium to each stock vial of cells. Resuspend the cells by gently pipetting using a 1000-μL pipette tip.

    Note: Up to 1 × 10⁷ single cells can be used per 1 mL of freezing media. For passages 1–4, freeze approximately 5 × 10⁵ cells per stock vial, whereas for passages 5 and beyond, use approximately 1 × 10⁶ cells per vial. Adjust the number of cells depending on the degree of organoid amplification.

  • e.
    Transfer the cell suspension to a cryovial. Place the cryovial in a cell-freezing container and freeze at −80°C.
  • f.
    On the following day, transfer the stock vials to a liquid nitrogen tank.
• 14.
  Thawing organoids.

  • a.
    Take the organoid stock vial from the liquid nitrogen tank and place in a dewar flask containing liquid nitrogen for thawing.
  • b.
    Place the cryovial in a floating tube rack and thaw in a 37°C water bath for 1 min, or until approximately half of the contents are thawed.
  • c.
    Meanwhile, label a 1.7-ml Eppendorf Tube with the information of thawing organoid¹⁴ and place it in a rack to prepare for use.
  • d.
    Add 1 mL of warm serum-free RPMI 1640 to the cryovial.
  • e.
    Resuspend by pipetting with a 1000-μL pipette tip approximately three times gently, until fully thawed.
  • f.
    Transfer the cell suspension to a prepared 1.7 mL Eppendorf Tube.
  • g.
    Centrifuge at 7,500 × g for 1 min at RT.
  • h.
    Remove the supernatant using a 1000-μL pipette tip and wash the cells with 1 mL of warm serum-free RPMI 1640.
  • i.
    Centrifuge at 360 × g for 5 min at RT.
  • j.
    Repeat steps 14-h to -i.
  • k.
    Remove the supernatant using a 1000-μL pipette tip.
  • l.
    Place BME aliquots on ice.
  • m.
    Resuspend the cell pellet with the required volume of BME (1 × 10⁵ cells/50 μL droplet), based on the number of viable cells at the time of freezing.

    Note: After freeze-thaw, more cells are seeded per dome to compensate for cell loss, so the BME volume is slightly increased to 50 μL per dome.

  • n.
    Place one droplet of the suspension in each well of the 24-well plate.
  • o.
    Incubate the 24-well plate at 37°C and 5% CO₂ for 30 min.
  • p.
    Add 500 μL of pre-warmed organoid growth medium containing 50 μM of Y-27632 to each well.
  • q.
    Incubate the plate at 37°C and 5% CO₂.
  • r.
    After 3 days, replace the growth medium without Y-27632.
• 15.
  Quality control for cryopreserved organoids.

  Note: This step is performed simultaneously with the start of Step 14. A portion of the thawed organoid suspension is seeded into a 96-well plate to assess cryopreservation quality control (QC), while the remaining cells are seeded into a 24-well plate for continued culture.

  • a.
    Seeding the organoids in a 96-well plate.

    • i.
      Perform the thawing steps, as described in steps 14-a to 14-m.
    • ii.
      Seed 2 μL of the BME-organoid suspension into three wells of a 96-well plate as triplicates using a 10 μL pipette tip.

      Note: Culture the remaining cell suspension as described in steps 14-n to 14-r.

    • iii.
      Following 2 min of incubation at RT, add 100 μL of warm organoid growth medium containing 50 μM of Y-27632.
    • iv.
      To prevent evaporation, add 200 μL of PBS to the outer wells surrounding the seeded wells.
    • v.
      Incubate the plate at 37°C and 5% CO₂.
    • vi.
      After 3 days, carefully remove the existing culture medium using a 200-μL pipette tip and replace it with organoid growth medium without Y-27632.
  • b.
    Organoid live/dead staining.

    • i.
      Prepare the 96-well plate containing organoids that have been seeded for 7 days.
    • ii.
      Following the instructions of the LIVE/DEAD Viability/Cytotoxicity Kit manual, prepare a staining solution by adding 0.5 μL of 4 mM Calcein AM and 1 μL of 2 mM Ethidium homodimer-1 to 1 mL of serum-free RPMI. Mix the solution thoroughly.
    • iii.
      Remove the cultures medium from the 96-well plate using a 200-μL pipette tip.
    • iv.
      Add 100 μL of the staining solution to each well.
    • v.
      Incubate the plate at RT for 30 min in the dark.
    • vi.
      Capture bright-field, green, and red fluorescence images using the Operetta CLS system, and measure the fluorescence intensity.
  • c.
    Data analysis.

    • i.
      Calculate cell viability using the following formula: $\frac{calceinAMfluorescencevalue}{CalceinAMfluorescencevalue+EthidiumHomodimer-1fluorescencevalue}\times 100(\%)$.
    • ii.
      Organoids with a viability of 60% or higher are considered acceptable (problem 9).

Drug response assessment using a 384-well plate

Timing: 1 h 30 min for plating; 3 days for cultures; 1 h for drug treatment; 5 days for cultures; 40 min for cell viability assay, as shown inFigure 9A

Note: This protocol is structured around a representative setup involving 14 drugs tested in a single 384-well plate. Each drug is assessed at 5 doses plus a no-treatment control (6 doses total), with 4 replicate wells per dose. The overall design is flexible and can be modified to accommodate different numbers of drugs, organoid lines, or plate layouts according to experimental needs (Figure 9B).

• 16.
  Preparing organoids for the drug response assay.

  • a.
    Repeat steps 12-a to 12-q to dissociate organoids into single cells and count the cells for seeding into 384-well plate.
  • b.
    Calculate the volume required to obtain 4 × 10⁵ cells (1,000 cells × 400 wells).

    Optional: To customize the experimental setup, define the following variables and calculate the total cell number as needed:

    • i.
      D = number of drugs
    • ii.
      C = number of dose conditions per drug (including control)
    • iii.
      R = number of replicate wells per dose
    • iv.
      N = number of cells per well (including 20% excess for handling loss)
      Then, calculate the total number of cells required using:
      Total cells = D × C × R × N.
      Refer to the layout in Figure 9B and adjust these parameters according to your experimental design.
  • c.
    Transfer this volume to a new 1.7-ml Eppendorf Tube.
• 17.
  Seed organoids into a 384-well plate.

  • a.
    Add 800 μL of BME to 7.2 mL of ice-cold organoid growth medium to achieve a 10% final concentration.
  • b.
    Mix the solution by vortexing, then place the mixture on ice.

    CRITICAL: To prevent premature solidification of BME, ensure that the BME and organoid growth medium are thoroughly chilled and maintained on ice from the moment of mixing until the seeding process is completed.

  • c.
    Centrifuge the tube containing the organoids at 360 × g for 5 min
  • d.
    Remove the supernatant.
  • e.
    Dispense 7 mL of 10% BME solution into a sterile reagent reservoir (SPL, Cat#21102, 23050).
  • f.
    Resuspend the organoid pellet in the remaining 1 mL of 10% BME solution using a 1000-μL pipette tip.
  • g.
    Transfer this suspension into the reservoir.
  • h.
    Resuspend and mix the solution thoroughly using a 10-mL serological pipette.
  • i.
    Using a multichannel pipette, dispense 20 μL of the suspension per well into all wells of the 384-well plate, excluding rows A and P.

    Note: For drug assays in 384-well plates, cells are resuspended in organoid growth medium containing 10% BME. This low-viscosity suspension does not form domes, enabling straightforward seeding by dispensing directly into the wells using a multichannel pipette. Dispense the suspension along the side wall of each well, close to the bottom, to prevent premature gelling. After seeding each plate, gently tap the bottom of the plate on the benchtop to help the suspension settle completely. Avoid tapping the sides of the plate, as this may cause the droplets to splash.

    CRITICAL: Gently agitate the reservoir before each aspiration to prevent organoid settling.

  • j.
    Fill each well in rows A and P of the 384-well plate with 80 μL of PBS to prevent evaporation.
  • k.
    Observe the plate under a microscope immediately after seeding to confirm even distribution of the organoid suspension.
  • l.
    Incubate the plate at 37°C with 5% CO₂ for 3 days.
• 18.
  Drug treatment of organoids.

  • a.
    Prepare the drugs by diluting them to various concentrations.

    Note: Since each well in the organoid plate already contains culture medium, prepare the drug solution at 2× the desired final concentration. When you add it in a 1:1 ratio, it will dilute to the correct concentration. One 96-well plate of 2× drug solution is enough to treat two 384-well plates of organoids (Figure 9B).

    • i.
      Prepare the organoid growth medium in a sterile reagent reservoir (SPL, Cat#23050).
    • ii.
      Using a multichannel pipette, dispense 225 μL of the organoid growth medium into rows A–G of columns 2–5 and 8–11 on a 96-well clear round bottom microplate.
    • iii.
      Take out the 100× stock vials of each drug from −80°C storage.
    • iv.
      Load 5 μL of the 100× drug solutions into each well from rows A–G of columns 1 and 12 of the 96-well plate for a total of 14 different drugs.

      Note: Adjust dilution factors if maximum solubility is lower than 100×.

    • v.
      Dispense 245 μL of the organoid growth medium into the wells containing the drugs.
    • vi.
      Perform serial dilutions by dispensing 25 μL sequentially from columns 1 to 5 and columns 12 to 8 using a multichannel pipette. Do not dilute columns 6 and 7.

      Note: Columns 6 and 7 can serve as controls for untreated, vehicle-only, or positive treatments.

  • b.
    Treat organoids with drugs.

    • i.
      On day 3, organoid formation and growth should be checked again before drug treatment.

      Note: This step is particularly important during initial optimization, for which image documentation is also recommended. Captured images using the Operetta CLS system on day 3 can support troubleshooting efforts by providing visual benchmarks for seeding consistency and growth patterns.

    • ii.
      Using a multichannel pipette, dispense 20 μL of the 2× diluted drug solution from each well of the 96-well plate into four wells of the 384-well plate.

      Note: Attach 200-μL tips to the seven channels of the automatic multichannel pipette. Set the pipette to dispense 20 μL per well in a 4-fold mode, and aspirate seven different drugs from the 96-well plate into each tip, starting with the lowest concentration.

    • iii.
      Incubate the plate at 37°C with 5% CO₂ for 5 days.
• 19.
  Assessment of organoid cell death induced by drugs.

  • a.
    Capture endpoint images of the organoid morphology in each well of the 384-well plate using the Operetta CLS system to assess drug-induced cytotoxic effects.
  • b.
    Perform the cell viability assay.

    • i.
      Thaw the aliquoted CellTiter-Glo 3D Reagent at 4°C.
    • ii.
      Incubate the CellTiter-Glo 3D Reagent at 22°C–25°C for approximately 30 min to equilibrate to RT.
    • iii.
      Add 20 μL of CellTiter-Glo 3D reagent to each well of the 384-well plate containing organoids.
    • iv.
      Wrap the 384-well plate in aluminum foil and incubate at RT for an additional 25 min.

      Note: Protect the plate from light during incubation to prevent potential degradation of the luminescent signal.

    • v.
      Gently shake the plate with a microplate shaker for 5 min at RT to induce cell lysis.
    • vi.
      Incubate the 384-well plate at RT for an additional 25 min.
    • vii.
      Measure the luminescence signal using a microplate reader (problem 10).
• 20.
  Data analysis of drug response assay.

  • a.
    Export raw luminescence readings obtained from the microplate reader to a spreadsheet
  • b.
    Assess data quality and reproducibility. Calculate the Z′-factor in Excel using the luminescence values from the no-treatment wells (negative control) and the wells treated with the highest drug concentration that induces complete cell death (positive control).

    Note: The Z′-factor can be calculated using the following formula: Z' = 1 – [3 × (SD_{negative control} + SD_{positive control}) / |Mean_{negative control} – Mean_{positive control}|] SD and Mean are the standard deviation and mean of the negative (no treatment) and positive control (100% inhibition) groups, respectively. A Z′-factor > 0.5 indicates a robust assay and is required to ensure the reliability of the results; only assays meeting this criterion should proceed to downstream analyses.

  • c.
    Normalize the luminescence values in each well to the average of untreated control wells (set as 100% viability).
  • d.
    Generate dose–response curves using GraphPad Prism.

    • i.
      In GraphPad Prism, go to New Table & Graph > XY, then select “Enter or import data into a new table”. choose “Numbers” for X, and “Enter replicate values in side-by-side subcolumns” for Y with 4 replicates.
    • ii.
      Click “Create” to open the data table.
    • iii.
      In the X column, enter the drug concentrations (preferably log-transformed) used in your experiment (e.g., log₁₀ of 0.1, 1, 10, 100, 1000 nM).
    • iv.
      In the Y columns, enter the cell viability values (%) for each concentration across 4 replicates.
    • v.
      After entering the data, click “Analyze” > “Dose-response – Inhibition” and select the model “log(inhibitor) vs. response – Variable slope (four parameters)”.
    • vi.
      Review the fitted curve and the Prism-calculated parameters, including IC₅₀ (half maximal inhibitory concentration).

      Note: IC₅₀ is useful for comparing the cytotoxic potency of different drugs. However, when the goal is to compare response patterns across organoid samples, it is more appropriate to analyze AUC values.

    • vii.
      Customize the graph for visualization and export.
  • e.
    Calculate the Area Under the Curve (AUC) in GraphPad Prism.

    • i.
      To obtain the Max AUC, create a reference sample in GraphPad Prism where cell viability is set to 100% across all tested concentrations
    • ii.
      Click the “Analyze” button from the top Data tab.
    • iii.
      Choose “Area under the curve” from the list of analysis types.
    • iv.
      Click OK to obtain AUC values
    • v.
      Export all AUC values to Excel and normalize each value using the following formula: $NormalizedAUC=\frac{AUC}{\mathit{Max}AUC}$

      Note: This normalization ensures that: Normalized AUC = 1 indicates no cytotoxic effect (full viability), Normalized AUC = 0 indicates complete loss of viability across all concentrations.

Figure 9.

Layout of organoid drug assay

(A) Day-by-day schedule of drug assay and brief experimental details.

(B) Design of a serial dilution plate for 14 types of drugs and a 384-well plate for drug treatment by concentration.

Expected outcomes

PDOs have emerged as valuable models for precision medicine, accurately reflecting the characteristics of the patient’s original tissue.¹⁴ This protocol enables the robust establishment and expansion of PDOs from a wide variety of specimen types, including surgically resected tissues, image-guided biopsies (e.g., EUS-FNB, PLB), and liquid biopsies (e.g., ascites and pleural effusions) (Figure 10). This broad applicability is particularly beneficial for patients with advanced-stage cancers who cannot undergo surgery, as it facilitates precision medicine through drug response evaluation of individual PDOs. Furthermore, these methods can be applied to organoid production using adjacent normal tissues derived from surgical specimens.

Figure 10.

Images of patient-derived specimens categorized by collection method

When seeded directly from clinical specimen, early-phase organoid cultures often contain heterogeneous cell populations including stromal cells, tissue fragments, immune cells, and RBCs (Figure 4). These non-tumor components are progressively eliminated over successive passages, yielding purified tumor organoids (Figure 7).

After passaging stabilized organoids, a single-cell state is observed, from which new organoids begin to form within 3 days, typically reaching ∼50 μm in diameter (Figure 11). Within 7–10 days, the average diameter of the organoids reaches approximately 150 μm. At this stage, organoid morphology can be categorized into three types: cystic, dense, and grape-like, depending on the cancer type, diagnosis, or genomic characteristics retained from the parent specimen (Figure 12).¹⁵ Passaging the organoids approximately every 2 weeks allows for up to five passages over a period of 3–4 months. From the third passage onward, one or more stock vials can be produced per passage.

Figure 11.

Sequential growth of PDOs over time

Growth of stabilized organoids after passaging on days 1, 3, 5, and 7. The scale bar represents 100 µm.

Figure 12.

Diverse morphologies of PDOs across different cancer types

PDOs from various cancer types exhibit diverse morphologies depending on the molecular characteristics. The method of specimen collection for each organoid is indicated in the top left corner, with abbreviations as follows: S for surgery, PB for punch biopsy, PLB for percutaneous liver biopsy, E for EUS-FNA, A for ascites, and P for pleural effusion. The scale bars represent 50 µm in the upper panels and 25 µm in the lower panels.

The organoid line is considered well-established and stable when it exhibits appropriate growth rates that allow for regular passaging and can be banked reliably with sufficient amplification of the tumor organoids. Therefore, a quality control experiment should be conducted to confirm whether they can be thawed and regrown for future experiments (Figure 13). These organoids can also undergo characterization through STR analysis, hematoxylin and eosin (H&E) staining, and immunohistochemistry (IHC) staining to assess their concordance with the original patient-derived specimens.

Figure 13.

Evaluating the quality of freeze-thawed organoids

To assess the feasibility of long-term cryopreservation and re-cultures of established organoid lines, frozen organoids are thawed and cultured for 7 days. Organoid viability and cell death is then evaluated. Live cells are stained with Calcein AM (green), while dead cells are stained with ethidium homodimer (Orange). Viability is quantified by measuring the fluorescence intensity of Calcein AM relative to the total fluorescence signal.

Additionally, the drug responsiveness of the organoids is evaluated through cell viability assays. This allows for the comparison of drug responses between different organoids and among various drugs for a single organoid, using metrics such as area under the curve (AUC) values derived from dose-response curves and z-scores.

These data can be used to identify drug sensitivities, define therapeutic windows, and prioritize candidates for further preclinical validation or personalized treatment strategies.

Altogether, the protocol not only standardizes the establishment and maintenance of organoids but also integrates downstream applications including cryopreservation, quality control, and drug screening into a single cohesive pipeline. Its flexibility and applicability to various specimen types enhance reproducibility and translational potential across laboratories. As such, it lays the groundwork for applying PDO models across diverse research and clinical contexts, from real-time therapeutic guidance to retrospective analyses of treatment response, thereby advancing a more integrated and biologically informed approach to cancer precision medicine.

Limitations

The success of establishing and culturing organoids using this protocol is largely dependent on the abundance of tumor cells present in the collected specimens. In cases such as oral cancer or specimens acquired via EUS-guided procedures, contamination by resident microbiota is common despite the use of high-grade antibiotics. These uncontrollable factors can significantly impact the outcomes of organoid cultures. Moreover, since niche factors vary among different cancer types, optimizing the growth medium to reflect these specific factors is essential. The successful establishment and amplification of organoids using this method require prior optimization of the organoid growth medium (refer to problem 2).

Troubleshooting

Problem 1

The tissue remains undigested after 1-hour enzymatic treatment. (Related Step: Step 2).

Potential solution

Do not extend digestion time beyond 1 hour, as prolonged exposure reduces cell viability. Tissue is highly fibrotic or dense, and enzyme penetration is limited. Mince as finely as possible in Step 2-b to increase the yield of living cells, rather than exposing the tissue to the dissociation enzyme for a long time. Proceed with culture even if undigested fragments remain; in some cases, organoids form more efficiently from intact tissue pieces.

Problem 2

While the establishment of organoids is observed, their growth is not sustained, ultimately leading to cell death. (Related Step: Step 10).

Potential solution

If the organoid growth medium is not adequately optimized, organoids may fail to grow or exhibit limited growth for up to 7 days post-seeding, with internal keratinization occurring.¹⁶ As such, the optimization of the culture medium is necessary, as described below. Determine the baseline medium using Advanced DMEM/F12 supplemented with HEPES, GlutaMAX, NAC, B-27, and essential growth factors, such as EGF, Noggin, and R-spondin 1. Subsequently, prepare test media by sequentially adding individual growth factors to the baseline medium to determine which factors enhance organoid growth. Prepare the BME-cell suspension and seed 2 μL of the mixture into each of three wells per test medium in a 96-well plate. Add 100 μL of each test medium to the wells and incubate. Observe the morphology, size, and density of the organoids closely during microscopy and change the medium every 2–3 days. If organoid growth is observed after more than 2 weeks, assess proliferation levels using a cell viability assay. Based on the results, identify the growth factors that enhance organoid growth and optimize the organoid growth medium accordingly.

Problem 3

The patient-derived cell suspension has low cell viability (Related Step: Step 10).

Potential solution

• •In the case of tissue specimens, prolonged enzyme treatment for complete dissociation can damage tumor cells, leading to reduced viability. Adjust the settings of the GentleMACS Octo Dissociator program based on tissue stiffness or reduce the dissociation time. Use 40 min as a standard protocol and modify the time as needed (from 15 min to a maximum of 1 h), considering the extent of tissue dissociation and tumor cell damage specific to the cancer type.
• •Generating and establishing organoids can be particularly challenging if the collected specimen is too small or lacks a sufficient number of tumor cells. This is a critical aspect of the process; thus, it is essential to either increase the quantity of the specimen collected or prioritize the collection of specimens that are rich in tumor cells whenever possible.
• •When the number of viable cells is low or the viability measured by trypan blue exclusion appears low, it is still recommended to proceed with dome seeding. In our experience, organoids can still form from samples with low viability, especially if they contain cancer stem cells. It is crucial to focus on creating a culture environment that supports the survival and proliferation of even a small number of viable tumor cells. To achieve this, red blood cells should be efficiently removed, and differentiated tissue debris minimized, as they may interfere with BME integrity and organoid formation.
• •When dealing with specimens containing high levels of RBCs or liquid biopsies with a significant number of immune cells, these smaller cells can be mistaken for dead cells, leading to an apparent reduction in viability. To address this, consider performing an additional RBC lysis step or using techniques like magnetic bead-based column sorting or fluorescence-activated cell sorting (FACS) to selectively isolate tumor cells before seeding. However, be aware that these methods may result in some loss of tumor cells.

Problem 4

Overgrowth of fibroblasts (Related Step: Step 10).

Potential solution

While cancer-associated fibroblasts (CAFs) can support tumor cell growth, they may pose challenges during the establishment of organoids. Fibroblasts often grow rapidly, adhere to the cultures surface, and expand within the basement membrane extract (BME), which can interfere with the efficient development of initial tumor organoids. Specimens with a high proportion of fibroblasts may also have a more complex tumor microenvironment, leading to a reduced number of tumor cells even if the specimen size is comparable. If co-growth of organoids and fibroblasts is detected, it is advisable to separate them during passaging and culture them independently. Typically, fibroblasts will adhere to the cultures plate surface, while organoids will grow within the BME, enabling efficient separation.

Problem 5

The embedded BME collapses within 1 week after seeding (Related Step: Step 10).

Potential solution

• •Inadequately dissociated tissue can cause the 3D dome to collapse. If the 3D structure collapses, passaging earlier than scheduled is necessary for its reconstruction and to ensure a supportive environment for growth. To prevent BME collapse, increase the solidification time after seeding to allow for adequate moisture evaporation. Alternatively, reduce the cell number or ensure even distribution of the tissue across droplets during seeding.
• •When embedding an excessive number of cells in a single droplet, the BME can easily disintegrate. It is advisable to reduce the volume of cells in each droplet, considering the amount of pellets and cell count.

Problem 6

When resuspending BME with the cell pellet, the BME tends to solidify and adhere to the walls of the pipette tip (Related Step: Step 3).

Potential solution

BME solidifies at RT and may harden upon contact with the pipette tip. To avoid this, store the pipette tips in the refrigerator and use them directly from cold storage. Additionally, to prevent exposure of BME to RT, the tube containing the BME-cell mixture should be maintained on ice throughout the entire seeding process. Coating the pipette tip with 0.1% BSA-PBS solution by aspirating and dispensing the solution before use also helps prevent cells from adhering to the tip walls.

Problem 7

During seeding, the BME resuspended cell solution spreads across the bottom of the plate instead of forming a 3D dome (Related Step: Step 3).

Potential solution

• •Thoroughly remove the supernatant to retain only the cell pellet before resuspending it in BME. If the BME is diluted, it will become less viscous, which can make droplet formation challenging.
• •Thaw the frozen BME at 4°C. Before use, ensure it is thoroughly mixed by vortexing or pipetting to achieve a uniform consistency. Inadequate mixing can cause the BME to be watery on the upper layer while the lower layer remains solid.
• •When seeding, carefully load the BME droplet at the center of the well. Droplets that touch the edges may cause the 3D structure to spread, affecting the integrity of the cultures.
• •Pre-warm the cultures plate to 37°C before seeding. After placing the BME droplet, invert the plate to create a hanging drop, and incubate it at 37°C. This method helps maintain the droplet’s shape and encourages proper organoid formation.

Problem 8

Cell pellet is not visible after organoid dissociation by TrypLE and centrifugation (Figure 14) (Related Step: Step 12).

Figure 14.

Appearance of a cell pellet embedded in partially dissolved BME

After incomplete dissolution of solidified BME, the cell pellet remained attached to BME fragments.

Potential solution

If there is no visible cell pellet at the bottom of the tube after centrifugation following organoid dissociation, this may be due to cells remaining clumped with partially degraded BME. To prevent this issue, after adding serum-free medium to dilute the TrypLE, invert the tube gently to mix thoroughly before centrifugation. If clumping persists, scrape the tube along the plastic compartment to physically dissociate the cells, then centrifuge again to check for the presence of a cell pellet. If clumps remain, extend the incubation time and repeat the dissociation process until the cells are properly separated.

Problem 9

Cryopreserved organoids show <60% viability upon thawing and QC (Related Step: Step 15).

Potential solution

• •If organoid subclones still grow beyond 100–150 μm in diameter, proceed with passaging as viable cells may still re-establish.
• •If most cells are non-viable and no organoids reach passage criteria, check whether freezing was performed according to protocol (e.g., controlled rate freezing, proper cryoprotectant use, prompt transfer to liquid nitrogen).
• •Consider that some organoid lines may not tolerate freeze–thaw well; in such cases, avoid cryopreservation or optimize freezing strategy specific to that line.

Problem 10

High standard deviation (SD) or coefficient of variation (CV) in drug response assay results using organoids (Related Step: Step 19).

Potential solution

Insufficient single-cell dissociation during organoid passaging can lead to uneven seeding in 384-well plates, with clumped cells potentially causing variations in cell number per well. To address this, ensure that cells are properly dissociated into a single-cell state without aggregation before proceeding. Observe the dissociation process under a microscope to confirm this. Additionally, when seeding cells using a reservoir, periodically mix the cell suspension to prevent cells from settling at the bottom or aggregating, ensuring consistent seeding across wells.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Yun-Hee Kim (sensia37@ncc.re.kr).

Technical contact

Technical questions on executing this protocol should be directed to and will be answered by the technical contact, Yun-Hee Kim (sensia37@ncc.re.kr).

Materials availability

Established PDOs are available for distribution from the Organoid Research Team at the National Cancer Center, Korea. Distribution of these materials is subject to a material transfer agreement (MTA).

Data and code availability

This protocol includes all datasets generated or analyzed during the study. No custom code was generated or used.

Acknowledgments

This study was supported by grants from the National Cancer Center, Korea (grant number: NCC-2210980 and NCC-2510700), and the National Research Foundation of Korea funded by the Korean government (MIST) (grant number: 2020M3A9A5036362).

Author contributions

S.K. drafted the original manuscript. S.K. and M.R.L. designed the study, performed the experiments, analyzed the data, and interpreted the results. M.R.L. revised and edited the manuscript. W.C. provided expert feedback. S.-Y.K. and Y.-H.K. secured funding, conceived the study, and supervised the project.

Declaration of interests

The authors declare no competing interests.