Multiparametric Tumor Organoid Drug Screening Using Widefield Live-Cell Imaging for Bulk and Single-Organoid Analysis

This protocol will allow researchers to not only streamline their analysis pipeline, but it will also allow them to extract more information out of their organoid stream, thereby enhancing the translational relevance. The main advantage of this protocol is it combines a streamlined workflow with an automated and clinically relevant analysis pipeline, thereby enabling researchers to obtain a compelling amount of information out of one single organoid screen. This technique can be used to predict clinical therapy response for cancer patients from these ex vivo organoid drug screenings with very promising early results.

The aim is to make the software solution platform agnostic so that researchers can use a live-cell imaging instrument that is available to them. Begin by enzymatically dissociating the patient-derived tumor organoids, or PDTOs, in the extracellular matrix, or ECM, domes in the six-well microplate. For this, first aspirate the medium, and wash the ECM domes once with PBS.

Then, add two milliliters of the dissociation enzyme in the well. Pipette the content up and down 10 times using a one-milliliter pipette to mechanically dissociate the organoids in the domes before incubating the plate for 10 minutes at 37 degrees Celsius. Post-incubation, pipette the content up and down, and check for the proper dissociation of the organoids to single cells using a microscope.

Collect the cell suspension in a 15-milliliter tube, and make the volume up to 10 milliliters by adding ADF+Centrifuge the tube at 450 g for five minutes at room temperature, and aspirate the supernatant using a Pasteur pipette and suction pump. Resuspend the pellet in 100 to 200 microliters of the full pancreatic ductal adenocarcinoma, or PDAC, organoid culture medium depending on the size of the pellet before counting the number of cells using a method of choice. To plate the single cells in ECM domes, dilute the appropriate amount of cell suspension by adding the required amount of thawed ECM.

Then, pipette up to 10 20-microliter droplets per well in a six-well plate preheated to 37 degrees Celsius. Invert the plate and incubate for 30 minutes at 37 degrees Celsius. Next, overlay the wells with the full medium supplemented with 10-micromolar Y-27632, and incubate for two to three days in an incubator.

To harvest the two-to three-day-old organoids, first aspirate the medium, and wash the organoids once with PBS. Then, add one to two milliliters of organoid-harvesting solution, pre-cooled to four degrees Celsius, to a well of the six-well plate depending on the number of ECM domes. Incubate the plate kept on ice, on a shaking platform for 10 minutes.

Now dissociate the ECM domes by pipetting the content in the wells up and down with a one-milliliter pipette. Once again, incubate the organoids for 10 minutes on ice before visually confirming the dissociation of the ECM domes under a microscope. Collect the organoids in a 15-milliliter tube, pre-coated with 0.1%BSA in PBS, and make up the volume to 10 milliliters by adding ADF+Centrifuge the tube at 200 g for five minutes at four degrees Celsius.

Next, aspirate the supernatant, and resuspend the pellet depending on its size in up to one milliliter of full PDAC organoid medium to the desired final organoid concentration. Open the pipetting robot control application, select Protocols, and click on Import before dragging and dropping the provided Supplementary File 6 into the designated field. Select the imported protocol.

And according to the layout shown in the Deck Setup field, place all the labware, including cooling elements and plasticware, in the decks. Use the left slot for the 25-milliliter reservoir and cooling element. Next, click on Run Protocol and Proceed to Setup.

Open the Labware Setup tab, click on Run Labware Position Check, and follow the instructions to calibrate the pipetting robot to the new hardware. Fill the 25-milliliter reservoir, placed on top of the cooling element, with the cooled organoid seeding solution, and click Start Run to put the pipetting robot in action. Next, centrifuge the microplate at a speed of 100 g for one minute at four degrees Celsius.

To create the drug dispensing protocol using the digital drug dispenser control software, hover over Plate 1 above the plate layout. Select Edit Plate Attributes, and fill in the plate type as 384 well, the additional volume as 50 microliters, and the DMSO limit as 1%Add fluids by clicking on the addition button next to Fluids, and double-click on the newly created fluid to name it. Then, select the class as DMSO based or aqueous plus Tween 20, and fill in the concentration.

To create the plate layout for drug titration, select Wells and click on Titration. For fluid, select the drug of interest, along with the desired highest concentration and lowest concentration. For replicates, select a minimum of two, and choose the desired titration pattern.

To create the plate layout for the positive control, select three wells, click on Set Value, and fill in two-micromolar staurosporine from a 10-millimolar stock in DMSO to induce maximum cell death. For Cytotox Green, select all the used wells, click on Set Value, and enter a value of 60 nanomolar per well. For the negative control and DMSO normalization, select all the wells with an additional four wells for the vehicle control.

Now right-click and select Normalization before assigning the normalized fluid class as DMSO based. Then, normalize to the highest class volume to obtain an equal DMSO concentration in each well. Next, click on the arrow under Run in the top left corner, and select Always Simulate, followed by clicking on Simulate to identify any errors and obtain the volumes of each drug to be prepared.

Now uncheck Always Simulate under the Run button before clicking on Run to start the drug dispensing protocol, and follow the instructions. Apply the sealing membrane to the microplate to prevent evaporation. Open the live-cell imager control software, and select Method Editor New before going to File to import the example method XML file.

Alternatively, follow the instructions in the manuscript to create a new file. Then, click on Start to initiate the scanning at T0.To merge and compress the data, create a new parent folder for transferring all the individual experiment folders generated for each scan at each time point by the live-cell imager control software, and name the respective folders by following the instructions given in the manuscript. Prepare an XLSX plate map in the digital drug dispenser control software by right-clicking on the plate map layout from the drug dispensing protocol and copying all the wells to paste the data in an XLSX file.

Remove Cytotox Green and staurosporine data, add a matrix for cell line, and enter a negative control and positive control. Open the Data Compression Tool. Click on Search, select the parent folder, and click on Compress to initiate image data compression.

All the TIFF image files are compressed into a single HDF5 for each well in a new datasets folder within the parental folder. To analyze the images, go to the image analysis web app platform, log in, and click on Add New Project in the Home tab. Enter the project name, press Continue, and select Add New Experiment before uploading the datasets folder containing the HDF5 files.

After uploading, go to the project and experiment folder to click on Upload Plate Map for additional functionalities. Then, sequentially click on Run Analysis, select Multi-Organoid Analysis, Default Parameters, and finally Analyze to initiate image analysis. Click on Download Results to download the raw data tables, which contain the measurements for each well and the segmented images or videos before confirming the accuracy of the analysis for further data processing.

To account for the variations in organoid seeding density and size, growth rate-based metrics were used to reduce variations between the replicates, as indicated by the reduced error bars. The images for the negative control, the positive control, and the combined gemcitabine and paclitaxel-treated PDTO showed that the therapy predominantly induced growth arrest with a limited amount of cell death, corresponding with normalized drug response, or NDR, value just below zero. Upon using two additional lines, PDAC_052 and PDAC_087, a clear difference in growth rate between the three lines was observed, supporting the use of GR metrics.

The NDR dose-response curves resulted in an increased dynamic range and separation between the three different patients, as compared to the GR curves. Determination of NDR over time showed that PDAC_052 and PDAC_060 had a very similar cytostatic drug response to a low dose of gemcitabine-paclitaxel, while a clear differential cytostatic versus cytotoxic response could be observed for the corresponding middle and high doses. These drug responses were consistent with the clinical responses observed in the patients.

The clonal dynamics revealed that PDAC_087 had the most resistant subclones following treatment, which is consistent with the aggressive nature of the disease observed in the patient, who, interestingly, was also the least sensitive to the positive control staurosporin. The most important aspect of the protocol is to keep all the solutions containing ECM cooled during the seeding because incorrect solidifying of the ECM can interfere with downstream image analysis. The automation of the wet lab work, the image analysis, and the downstream data processing can be helpful to accelerate the research and identify novel combination therapies more easily.