We present a protocol that can be used to conduct therapeutic drug testing with patient-derived ovarian cancer organoids.
Ovarian cancer is a fatal gynecologic cancer and the fifth leading cause of cancer death among women in the United States. Developing new drug treatments is crucial to advancing healthcare and improving patient outcomes. Organoids are in-vitro three-dimensional multicellular miniature organs. Patient-derived organoid (PDO) models of ovarian cancer may be optimal for drug screening because they more accurately recapitulate tissues of interest than two-dimensional cell culture models and are inexpensive compared to patient-derived xenografts. In addition, ovarian cancer PDOs mimic the variable tumor microenvironment and genetic background typically observed in ovarian cancer. Here, a method is described that can be used to test conventional and novel drugs on PDOs derived from ovarian cancer tissue and ascites. A luminescence-based adenosine triphosphate (ATP) assay is used to measure viability, growth rate, and drug sensitivity. Drug screens in PDOs can be completed in 7-10 days, depending on the rate of organoid formation and drug treatments.
Although rare, ovarian cancer is one of the most lethal gynecological cancers1,2. A challenge in developing new treatments is that ovarian cancer is heterogeneous, and the tumor microenvironment differs greatly among patients. Additionally, many ovarian cancers develop resistance to platinum-based chemotherapy and poly (ADP-ribose) polymerase inhibitors, highlighting the need for greater therapeutic options3,4,5.
One approach that may be useful in identifying new therapeutics is using patient-derived organoids (PDOs). Organoids are three-dimensional clusters of multiple cell types that self-organize and form in vitro "mini-organs"6,7,8,9,10. Organoids can recapitulate important tissue morphology and gene expression profiles11,12. Some of the first organoids were derived from intestinal, gastric, and colon cancer cells from both mice and humans8,9,13. Long-lived organoid cultures have been established from a wide range of benign and malignant tissues, including the bladder, colon, stomach, pancreas, brain, retina, and liver14,15,16. We previously demonstrated methods to establish PDOs from ovarian cancer tumors and ascites samples17. PDOs can be used to study molecular characteristics, cellular mechanisms, and novel drug treatments18,19,20. PDOs have several advantages over traditional two-dimensional primary cell cultures for drug screening. Although primary two-dimensional cultures are a low-cost method for drug screens, primary cell cultures are single-cell types and lack the three-dimensional architecture of tumors21,22,23. Nevertheless, PDOs are a precious resource, and cost-effective protocols are needed to optimize their use in therapeutic drug screening.
This article describes an in vitro method to use ovarian cancer PDOs to test the effects of known or candidate drugs. Whereas current medium- and high-throughput drug screens using PDOs require expensive automated dispensing instruments24,25,26, this cost-effective method uses readily available basic lab supplies and an ATP-based cell viability assay in a standard 96-well plate format (Figure 1A). This method will facilitate preliminary tests of novel ovarian cancer drugs prior to scaling up to larger screens27,28. Although ovarian cancer PDOs are used here, this method can be applied to other cancer organoid models.
The collection of human specimens for this research was approved by the Washington University School of Medicine Institutional Review Board. All eligible patients over the age of 18 years had a diagnosis or presumed diagnosis of high-grade serous ovarian cancer and were willing and able to provide informed consent. The tumor tissue from either primary or metastatic sites, in addition to ascites and pleural fluid, were obtained from consented patients at the time of care.
1. Selection of established PDOs for viability assay
NOTE: Typically, these organoids are within the first five passages. As described previously, ovarian cancer PDOs are formed by resuspending primary cell suspension in basement membrane extract (BME) such as Cultrex or Matrigel17 (see Table of Materials).
2. Reagent preparation for the viability assay
3. Plating of organoids (~Day -2)
NOTE: This step must be performed 1-3 days before adding the drugs. Before beginning, warm all reagents (Base Media, Advance Organoid Media, and organoid dissociation reagent; see Table of Materials) to 37 °C in a water bath. Thaw BME in an ice water bath.
4. Addition of drugs for the viability assay on Day 0
NOTE: Day 0 refers to the day the drugs are added to the fully formed organoids.
5. Ending the viability assay for readout (~Day 7)
NOTE: This step can be performed on Days 7-10. The assay length must be determined according to the half-life and pharmacodynamics of the drugs being tested.
6. Data analysis
These results illustrate the response of two PDOs to the chemotherapy drug carboplatin, which is used to treat ovarian cancer. Organoids were derived from tumor biopsy (PDO #1) and from ascites (PDO #2). These organoids were selected based on their perceived doubling time (1-2 days) and morphological appearance (formation of many large organoids). Both PDO #1 and PDO #2 were plated on Day -2, at passage two, and carboplatin was added on Day 0. We tested the following carboplatin concentrations diluted in Advance Organoid Media: 1, 5, 10, 25, 50, and 75 µM. At the conclusion of the experiment on Day 7, the viability assay reagent was added to the plate, and the results were analyzed. Figure 2A depicts the percentage of live cells after carboplatin treatment.
Next, the online GR Calculator was used to analyze the data. In the absence of cell count data, the luminescence values measured in the viability assay were used. After exporting the GR metrics, the GR values were graphed, which are the ratios between perceived growth rates in treated and untreated conditions normalized to a single cell division34. These values were then plotted against the carboplatin concentrations (Figure 2B). Table 1 summarizes the GR metrics, including the respective control and treated cell doubling times and the GR Area Over the Curve (GRAOC), which integrates the dose-response curve over a range of carboplatin concentrations tested34. Sensitivity to a particular drug can be determined by interpreting the GR50 value, corresponding to the concentration at which the drug has a half-maximal effect. For example, the GR50 value for PDO #1 is much higher than that for PDO #2 (4.85 µMvs. 0.97 µM), indicating that PDO #1 is more resistant to platinum chemotherapy than PDO #2.
Figure 1: Patient-derived organoids before drug screening. (A) Experimental outline for PDO drug screen. Given the doubling time of the PDO line and the drug exposure time, the experimental plan may need to be adjusted. (B) Representative brightfield images (40x) of two ovarian cancer PDO lines (#1 and #2). Scale bar = 50 µm. Abbreviations: PDO = patient-derived organoids. Please click here to view a larger version of this figure.
Figure 2: Representative results of ovarian cancer PDO following carboplatin treatment. (A) Two PDO lines were treated with increasing concentrations of carboplatin for seven days. The X-axis shows carboplatin concentration; the Y-axis displays the % of live cells normalized to control organoids (no carboplatin). Assays were completed in triplicate with two biological replicates. The error bars denote the standard deviation. (B) GR-value graph representing the logarithmic carboplatin concentration (x-Axis) and the GR value (Y-Axis). These values were generated in the online GR calculator. The error bars indicate the standard deviation. Please click here to view a larger version of this figure.
Treatment | Control Cell Doubling Time | Treated Cell Doubling Time | GR50 | GR_AOC | |
PDO #1 | Carboplatin | 0.744 | 0.112 | 4.85 | 1.28 |
PDO #2 | Carboplatin | 0.972 | 0.0532 | 0.97 | 1.51 |
Table 1: Table depicting the control cell doubling time, treated cell doubling time, GR50, and GR_AOC values generated in the GR calculator. Abbreviations: PDO = patient-derived organoids, GR = growth rate, GR_AOC = growth rate area over the curve.
This article describes a method that can be used to assess the therapeutic effects of conventional or novel drugs on ovarian cancer PDOs. Researchers must consider several issues before conducting the viability assay in the PDO model.
First, when selecting a PDO to use in the viability assay, one must determine the ideal organoid type (tumor vs. ascites) and passage number for their needs. In our experience, ascites-derived PDOs grow more rapidly and are easier to generate than tumor-derived PDOs. Because this assay depends on the growth rate, using organoids that take a long time to form/grow may be difficult. We have only successfully used PDO lines with doubling times of less than 4 days.
Second, opaque black 96-well plates were used in this study. Since the ATP viability assay is luminescence-based, transparent wells will reduce the signal intensity and generate signal contamination. Opaque-walled, transparent bottom plates allow for the visualization of organoids during the assay and may help monitor treatment-induced changes in cell morphology. Although it is a strength related to cost, a limitation of this assay is using a 96-well plate, which limits the number of samples and drugs that can be evaluated simultaneously.
Third, the number of cells plated must be carefully considered and optimized for viability assays. This is especially true because of the variable growth rate of PDO lines; too few cells will not form organoids, and too many cells will lead to organoid overgrowth. To ensure the uniform distribution of cells, an automatic cell counter was used and maintained the same BME-to-media ratio (75:25). This high percentage of BME ensures the droplets remain solidified for the entirety of the assay. Here, 3 µL droplets of BME were placed in the center of the well. Although the size of the droplet can be increased, we caution against coating the whole well. Coating the whole well will cause the organoids to settle in the edges of the well, which will both impede their overall growth and affect the viability assay results. Off-center placement of the droplet is fine as long as it does not touch the edges of the well.
Fourth, self-pipetting introduces human error, but this can be overcome by attention to detail and the inclusion of additional control wells.
Finally, the length of the assay must be carefully selected. Prolonged exposure to drugs will affect the viability of PDOs, independent of the drug mechanism of action. For this reason, it is important to test a range of drug concentrations over a minimum of five days. It is important to decide whether the media will need to be changed for longer periods because reduced levels of growth factors will impede the effects of the drugs36. To examine whether the media will need to be changed during the assay, one needs to compare the results from Day 0 and Day 7 controls. Untreated PDO controls should continue to grow throughout the assay continuously.
As PDOs continue to advance in complexity and better recapitulate their tissues of origin, their use should improve drug discovery. However, PDOs are likely to remain a precious resource and require cost-effective methods to preserve and optimize their use. Unlike medium- and high-throughput techniques, this protocol can be used to test known and novel compounds at lower cost with readily available materials and equipment. Finally, this method can be readily adapted to different cancer organoid models.
The authors have nothing to disclose.
Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA243511. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors thank Deborah Frank for her editorial comments.
1.5 mL Plastic Tubes | |||
15 mL Plastic Tubes | |||
96 well Flat Black Plates | MidSci | 781968 | |
Advance Organoid Media | see Graham et al 2022 (Jove) | ||
Advanced DMEM/F12 | Thermo Fisher | 12634028 | |
Automated Cell Counter | Thermo Fisher | AMQAX1000 | |
Brightfield Microscope | |||
Carboplatin | Teva Pharmaceuticals USA | NDC 00703-4246-01 | |
CellTiter-Glo 3D Viability | Promega | G9681 | |
Cultrex | R & D Systems | 3533-010-02 | |
DMSO | Sigma Aldrich | D2650-100ML | |
Glutamax | Life Technologies | 35050061 | |
GR Calculator | http://www.grcalculator.org | Online calculator | |
GraphPad Prism | GraphPad Software, Inc. | ||
HEPES | Life Technologies | 15630080 | |
Matrigel | Corning | 354230 | |
Microsoft Excel | Microsoft | ||
Penicillin-Streptomycin | Thermo Fisher | 15140122 | |
Plate Rocker | |||
Sterile P10, P200, and P1000 Barrier Sterile Pipette Tips | |||
Sterile P10, P200, and P1000 Pipettes | |||
Tecan Infinte 200Pro Plate Reader; i-Control Software | Tecan | ||
TrypLE | Thermo Fisher | 12605010 | Organoid dissociation reagent |