The mouse pancreatic tumor organoids were derived from flank tumors developed in c57bl/6 mice following subcutaneous injection of Panc02 cells. Organoids were grown and cultured in Matrigel^® domes on 24-well plates, as described previously (3).

The mouse pancreatic tumor organoids were generated and propagated in Matrigel^® domes in 24-well plates as described above. Immunofluorescence staining was performed as follows: the organoids were fixed in formalin tissue fixation buffer solution with 10% neutral buffer (Sigma Chemical, United States). Before staining, the fixative was removed, and organoids were washed with 1× PBS three times for 5 min each. Organoids were then permeabilized with 0.2% Triton X-100 for 5 min, washed with 1× PBS for 5 min, and blocked with the blocking buffer (Vector Labs, United States) for 1 h. Primary antibodies used were α-SMA (smooth muscle actin): rabbit polyclonal unconjugated 1:100 dilution (ABclonal antibody, #A7248, United States), vimentin: Alexa Fluor 594 conjugated anti-mouse vimentin 1:300 dilution (BioLegend, #699303, United States), and γ-H2AX: phospho gamma S139 H2AX-anti-rabbit (Cell Signaling Technology, #9718S, United States) 1:400 dilution. Incubation with primary antibody was performed overnight in a 1:5 diluted blocking buffer in a 4 °C cold room. Next, the primary antibodies were removed and washed as described above. For the unconjugated α-SMA and γ-H2AX detection, Alexa Fluor 488 rabbit secondary antibody (Thermo Fisher Scientific, #A11008, United States) at 1:300 dilution was used for 1 h. Hoechst 33342 dye (Thermo Fisher Scientific, #H21492, United States) with a concentration of 1 μg/mL was then added for 5–10 min. The antibody and Hoechst dye were then discarded, and the organoids were washed in the same way as described earlier. Finally, the organoids were kept in 1× PBS, and images were captured using an EVOS fluorescence microscope at 20x magnification (Thermo Fisher Scientific, United States). For staining, slides were deparaffinized, unmasked (Vector Labs, United States), and blocked with blocking buffer (Vector Labs, United States). Staining with the α-SMA primary antibody was the same as described above. For the secondary antibody, Vectastain Universal Quick Kit was used according to the manufacturer’s instructions. Finally, ImmPACT DAB (Vecta Stain, United States) was used to develop a brown color. After washing with water, the slides were air-dried and mounted with cytoseal-60 (Epredia, United States). The images were captured with an EVOS Xl color microscope.

The organoids were grown in 25 μL Matrigel^® domes on 24-well plates with 500 μL of organoid growth media. Treatment groups included 5-FU (0–100 μM), gemcitabine (5 μM), RT (4 Gy or 8 Gy), and 5-FU or gemcitabine + RT. For the single-modality groups, treatment was administered once, and organoids were allowed to grow for 5 days. For the combination groups, RT was administered 24 h after the initiation of chemotherapy treatment, and organoids were allowed to grow for an additional 4 days. At the end of the treatment, images were captured with an EVOS light microscope and fixed in formalin, as described above. After removal of the fixative, the organoids were washed with 1x PBS, three times each for 5 min, permeabilized with 0.2% Triton X-100 for 5 min, and stained with Hoechst dye (1 μg/mL) for 10 min. The images were captured with an EVOS fluorescence microscope (blue filter).

Organoids were grown and treated as above. At the end of radiation treatment, organoids were incubated with ROS substrate DHE (dihydroethidium, Medchem Express, United States) 20 μg/mL in growth media and incubated for approximately 2 h. Images were captured with the EVOS fluorescence microscope (red filter). Brightfield images were also obtained.

Tumor organoid response after treatment was assessed by averaging the size of the five largest organoids, as determined by ImageJ. The organoids that did not receive treatment were considered a control group, and all groups’ average sizes were normalized to the average size of the control. The responses of the groups were directly compared using a one-sided Student’s t-test, with a particular focus on comparing groups receiving combination treatments to those receiving the most effective individual component of those combinations. Differences in the average organoid size were considered significant with a p-value of <0.05.

It has been shown that mouse PTOs, derived from pancreas tumor tissues or patient-derived xenografts, closely mimic many of the histological, genetic, and phenotypic features of PDAC (9, 10). In this study, we have observed that tumor organoids may maintain the key characteristics of the original TME, expressing α-SMA and vimentin in patterns similar to tumor tissue (Figure 2). Thus, these tumor organoids represent cellular heterogeneity, tumor architecture, and stromal interactions with other important cell types, including but not limited to cancer-associated fibroblasts (CAFs).

The sensitivity of mouse PTOs to different doses of 5-fluorouracil (5-FU), gemcitabine, and radiation as monotherapy and combination therapy was evaluated. Tumor organoids were treated with 10–100 μM of 5-FU and monitored for growth inhibition by brightfield imaging. The data showed dose-dependent growth inhibition. Only 100 μM of 5-FU achieved >50% inhibition of tumor organoids, suggesting that single-modality treatment needs high doses to inhibit tumor organoid growth. Notably, co-treatment of 5-FU with radiation (e.g., 25 μM + 8 Gy) yielded synergistic effects and showed significant growth inhibition (p < 0.05) compared to single-modality treatments (Figure 3). Furthermore, we observed >70% growth inhibition in tumor organoids treated with 100 μM of 5-FU combined with 8 Gy of radiation. It is also important to note that the combination of 50 μM 5-FU + 8 Gy RT resulted in a response that was not consistent with a dose-dependent trend, which might be expected given the responses to combinations of 5-FU with 4 Gy RT. This is most likely due to the presence of an outlier organoid, which was approximately 2 standard deviations larger than the average of the other organoids in the group and was the largest in any group receiving 8 Gy of RT. Given the totality of the evidence, it is reasonable to conclude that a combination of low doses of 5-FU + radiation reduced organoid viability more effectively than either treatment alone.

The response of PTOs to gemcitabine + radiation treatment was also examined. Tumor organoids treated with a low dose of gemcitabine followed by 4 Gy or 8 Gy of radiation showed greater growth inhibition compared to single-modality treatments. There was approximately 50 and 70% growth inhibition in PTOs treated with 4 Gy + 5 μM gemcitabine and 8 Gy + 5 μM gemcitabine, respectively (Figure 4). While the response of the group treated with 4 Gy + 5 μM gemcitabine was not significant compared to the response of the group treated with gemcitabine alone (p = 0.10), it was significant when compared to the group treated with 4 Gy alone (p < 0.05). Additionally, immunofluorescent staining revealed a disorganized expression of vimentin following treatment with 4 Gy + 5 μM gemcitabine, which was not observed following treatment with either modality in isolation (Figure 5). These data suggest that the combination treatment was more effective, leading to enhanced PTO growth inhibition and TME disruption.

This study demonstrated that pancreatic tumor organoids exhibit distinct responses to individual and combination treatments. Specifically, the combination treatment of low doses of 5-FU or gemcitabine with radiation shows significant inhibition of tumor organoid growth. Therefore, it is reasonable to consider how the combination treatment approach may target different signaling pathways to exert a synergistic effect and overcome therapy resistance.

Radiation and chemotherapy drugs such as 5-fluorouracil and gemcitabine each use cytotoxic effects through different mechanisms. When used alone or in combination, these therapies target various cellular signaling pathways, leading to cell death via DNA damage, cell cycle arrest, and replication stress (15). When PTOs were treated with either gemcitabine or radiation, cytotoxic ROS were produced; however, the combination of gemcitabine and radiation appeared to generate a greater amount of ROS (Figure 6).

We also assessed the presence of double-strand breaks (DSBs) in tumor organoids treated with gemcitabine and radiation individually and in combination by staining for γH2AX. After treatment, H2AX was quickly phosphorylated at the site of DSBs, which was detected using immunofluorescence microscopy. The data also demonstrated that gemcitabine acts as a radiosensitizer, increasing radiation-induced DNA DSBs (Figure 7). This finding would be consistent with pancreatic cancer cell line work suggesting that gemcitabine synchronizes cells in the S-phase and inhibits repair pathways, making them more vulnerable to radiation-induced DSBs (16).

Based on these results, we envisage that 5-FU and gemcitabine both reduce the DNA repair ability of cells and, as a result, cause tumor cells to become more sensitive to radiation-induced DNA strand breaks and cell death.