1. Ethic issue
This study was conducted after the proposal had been approved by the medical ethics committee of Guangdong Provincial People's Hospital, Guangzhou, China (No. KY-Q-2021-97029-3). All procedures were carried out in accordance with the Helsinki Declaration. The sources of the BC samples are shown in Table 1.
2. Sample collection and organoid culture
Fresh BC tissue or metastatic samples were washed 3-5 times in PBS (Gibco, NY, USA) containing 500 U/mL Penicillin (Gibco, NY, USA), 500 mg / mL Streptomycin (Gibco, NY, USA) and 50 mg / mL Nystatin (Sangon Biotech, Shanghai, China). The tumor tissues were cut into small pieces (about 0.5-1 mm in diameter) and incubated with 1 ml TrypLE (Gibco, NY, USA) at 37 ° C for 45 minutes. After digestion, the mixture was used to remove tissue fragments through a 70 mm cell filter, centrifuged for 300 rpm for 5 minutes, and then the isolated cells were re-suspended in standard BC culture medium and mixed with a 1:2 volume of Matrigel matrix (Corning, NY, USA). 20 ml droplets were added to the preheated 48-well plate. After the droplets were solidified, 200 ml of the standard ABCO medium was added to each well and incubated under 37 ℃, 5% CO2 and 90% humidity condition. The medium was renewed in two-day intervals. The standard BC organoid culture medium was composed of heparin (Stemcell technologies, Vancouver, Canada), cortisone (Stemcell technologies, Vancouver, Canada) and mammocult ™ Human Medium KIT (Stemcell technologies, Vancouver, Canada), 100 U / ml penicillin (Gibco, NY, USA), 100 mg / ml streptomycin (Gibco, NY, USA), 20 mg / ml of nystatin ((Sangon Biotech, Shanghai, China).
3. Histological and immunohistochemical staining
In order to show the three-dimensional structure of ABCO, 1ml cell recovery solution (Corning, NY, USA) was added to the ABCO containing well and incubated at 4℃ for 1h. After being diluted with 1ml/well cold PBS, the solution was centrifuged (4000 rpm, 4 min, 4 ℃) and then fixed in 4% paraformaldehyde for 1h. The fixed ABCOs were mixed with 2% warm agarose and embedded in paraffin. Paraffin sections and standard HE staining were performed on all ABCOs and tissues. The antibodies used were as follows: mouse ant- Bcl-2 (Proteintech, Chicago, USA), rabbit ant-Ki67 (Abcam, Cambridge, MA, USA), rabbit ant-pSTAT3 (Abcam, Cambridge, MA, USA) , rabbit ant-ER (Abcam, Cambridge, MA, USA), rabbit ant-PR(Abcam, Cambridge, MA, USA), rabbit ant-HER2（Abcam, Cambridge, MA, USA), Coralite488-conjugated affinipure goat anti-rabbit IgG (Proteintech, Chicago, USA) and Coralite594-conjugated goat anti-mouse IgG (Proteintech, Chicago, USA). The experimental procedures of immunofluorescence and immunohistochemistry refer to the previously published experimental scheme. The images were collected with Nikon immunofluorescent microscope (Nikon, ECLIPSE Ci-L, Tokyo,Japan).
4. Drug treatment
The ABCOs were cultured in ABCO medium containing 100 mM resveratrol (Sigma-Aldrich, USA) , 1 mg/ml paclitaxel (MedChemExpress, NJ,USA) , 1mM gemcitabine (MedChemExpress, NJ,USA)  or 1 mM fulvestrant (MedChemExpress, NJ,USA) , respectively. The treatments were last for 96h and the drug-containing medium was changed every 48 hours. After 96h of the treatments, the ABCOs were analyzed by cell viability/death test and EdU cell proliferation assay. The drug use scheme of drug sensitivity test from each patient is provided by clinicians to promote the principle of individualized treatment.
5. Cell viability assay
Cell viability was elucidated by trypan blue (Invitrogen, Carlsbad, CA, USA) discrimination staining and Calcein-AM/PI double staining kit (Beyotime, Shanghai, China) according to manufacture’ instructions. Briefly, the ABCO cells were stained with 0.4% trypan blue, and the stained nonviable cells and unstained viable cells were counted with blood cell counting plate and cell death rate was determined by the stained cell number/the total cell number. In Calcein-AM/PI double staining test, the living cells with intact membrane and esterase activity are distinguished by strong and uniform green fluorescence produced by enzymatic hydrolysis of Calcein-AM, while the dead cells with damaged membrane emit bright red fluorescence due to the combination of propidium iodide and nucleic acid. The cell labeling patterns of ABCO population were captured under fluorescent microscope (Nikon, ECLIPSE Ni-U, Tokyo, Japan), followed by calculation of the dead cell fraction. In order to compare the cell death rate caused by the same drug in ABCO populations from different patients, adjusted mortality (%) was calculated via removing the natural mortality of ABCOs (the death rates of normally cultured ABCOs) from the total one by the formular shown below. All the experiments were repeated for three times. The mean adjusted mortality was the mean of the adjusted mortality over three replicates.
Adjusted mortality (%) = (treatment mortality - control mortality) / (1 - control mortality) * 100%
6. Cell proliferation assay
For EdU cell proliferation assay, the ABCOs were cultured in 48 well plates and then treated with different anticancer drugs selected for test for 96 hours. The proliferation activity of ABCOs was determined by EdU Kit (Beyotime, Shanghai, China), following the manufacturer’s instruction. Briefly, the ABCOs in each group were labeled with 5-acetyl-2 '-deoxyuracil (EdU) for 8h, and then incubated with Click additive solution for 30 minutes in darkness at room temperature. 5 mg/ml Hoechst 33342 (Beyotime, Shanghai, China) was used for nuclear contrast staining. The images of Hoechst 33342 and EdU staining were captured by fluorescence microscope (Nikon, ECLIPSE Ni-U, Tokyo, Japan) respectively and merged thereafter.
7. Determination of intracellular reactive oxygen species (ROS）levels
2 '- 7' - dichlorodihydrofluorescein diacetate (DCFH-DA) is a non fluorescent intracellular compound, which is cleaved by cytolactonase and then converted to fluorescent compound under the oxidation of ROS. Intracellular ROS levels were measured by hydrogen peroxide dependent DCFH-DA oxidation (Beyotime, Shanghai, China). ABCOs were treated with 100 mM Res for 0h, 2h, 4h and 6h, and washed at each time point by DMEM/F12 (Gibco, NY, USA). The ABCOs were incubated with 10 mM DCFH-DA in DMEM/F12 (Gibco, NY, USA) for 20 min at 37°C in the darkness, followed by three washes with DMEM/F12. The ABCO-bearing coverslips were collected at 0, 2, 4 and 6 hour time points and stained with DCFH-DA in situ. The cells were observed and photographed with a fluorescence microscope (Nikon, ECLIPSE Ni-U, Tokyo, Japan).
8. Statistical analyses
All the experiments were repeated at least three times. Calcein/PI test data were analyzed by Student’s t-test. Chi-square test was used to analyze the relationship between the efficacy of Res and the expression of pSTAT3.The data were analyzed by SPSS software (version 26.0; SPSS, Chicago, IL, USA). If necessary, P values are illustrated in the figure legend.