Construction of gastric cancer patient-derived organoids and their utilization in a comparative study of clinically used paclitaxel nanoformulations

Abstract

Background: Gastric cancer (GC) is a highly heterogeneous disease with many different histological and molecular subtypes. Due to their reduced systemic adverse effects, nanoformulation agents have attracted increasing attention for use in the treatment of GC patients in the clinic. To improve therapeutic outcomes, it is vitally necessary to provide individual medication references and guidance for use of these nanoformulations, and patient-derived organoids (PDOs) are promising models through which to achieve this goal.

Results: Using an improved enzymatic digestion process, we succeeded in constructing GC PDOs from surgically resected tumor tissues and endoscopic biopsies from GC patients; these PDOs closely recapitulated the histopathological and genomic features of the corresponding primary tumors. Next, we chose two representative paclitaxel (PTX) nanoformulations for comparative study and found that liposomal PTX outperformed albumin-bound PTX in killing GC PDOs at both the transcriptome and cellular levels. Our results further showed that the different distributions of liposomal PTX and albumin-bound PTX in PDOs played an essential role in the distinct mechanisms through which they kill PDOs. Finally, we constructed patient-derived xenografts model in which we verified the above distinct therapeutic outcomes via an intratumoral administration route.

Conclusions: This study demonstrates that GC PDOs are reliable tools for predicting nanoformulation efficacy.

Keywords: Albumin-bound paclitaxel; Gastric cancer; Liposomal paclitaxel; Nanoformulation; Patient-derived organoids.

Fig. 1

Construction and histopathological characterizations of GC PDOs. A Schematic overview of GC PDOs isolation, culture and histopathological analysis. B Number of passages and freeze-thaw status of GC PDOs derived from ten patients (numbered GC1 to GC10). Each block indicates one passage or a freeze-thaw cycle. For example, GC1 PDOs were passaged more than ten times and successfully thawed after freezing at passage 12, whereas GC9 PDOs stopped growing at passage 2. C Representative bright-field images of GC4 PDOs showing their morphological features during isolation, culture and passage. Scale bar, 100 μm. D Representative bright-field images and HE staining images of GC PDOs and the primary tumors from which they were derived. These tumors (GC5, GC2, and GC3) represent three histological subtypes. Scale bar, 50 μm. E Representative images showing IHC staining and AB staining of GC3 PDOs and the primary tumor. Scale bar, 30 μm.

Fig. 2

Genomic comparisons of GC PDOs with the primary tumor from patient GC3. A Heatmap showing CNVs in the primary tumor and the corresponding GC PDOs. The columns represent genomic positions from chromosomes 1 to 22, and the colors in the plot correspond to the estimated log₂ copy ratios of the genomic regions. B CNVs heatmap of GC driver genes in the primary tumor and the corresponding GC PDOs. Gene copy numbers are transformed as log₂ ratios per gene. C Heatmap showing gene mutation variations in the most frequently mutated GC genes. SNV: single-nucleotide variant. D Bar graphs showing the point mutation types found in the primary tumor and in the corresponding GC PDOs.

Fig. 3

Response of GC PDOs to clinically used PTX nanoformulations. A Schematic overview of the drug sensitivity analysis protocol. B Size distributions, zeta potentials, and representative TEM images of Albu-PTX and Lipo-PTX. C–J Representative bright-field images and dose-response kill curves of various PDO lines treated with Albu-PTX or Lipo-PTX. PDOs were treated with PTX at concentrations ranging from 1.5 × 10^− 4 µM to10 µM for 5 days. Scale bar, 200 μm. K IC₅₀ values for various PDO lines. Data in (B–J) are presented as the means ± s.d (n = 3)

Fig. 4

Distinct cytotoxic effects of PTX nanoformulations on GC1 PDOs at the transcriptome and cellular levels. A Transcriptome analysis of GC1 PDOs after treatment with Albu-PTX or Lipo-PTX. Differential gene cluster analyses are shown as heatmap. B KEGG pathway analysis of GC1 after treatment with Albu-PTX or Lipo-PTX. C GSEA of GC1 for gene sets that were changed in the Lipo-PTX treatment group versus the Albu-PTX treatment group. D Live-Dead staining analysis of GC1 PDOs after treatment with Albu-PTX or Lipo-PTX. Green: live cells; red: dead cells. Scale bar, 50 μm

Fig. 5

Distinct spatiotemporal distributions of PTX nanoformulations in GC1 PDOs. A The distributions of PTX nanoformulations in GC1 PDOs as revealed by CLSM at the indicated time points. Blue: DAPI; red: fluorescent-labeled Albu-PTX or Lipo-PTX. Scale bar, 50 μm. B The distributions of PTX nanoformulations in GC1 PDOs as revealed by fluorescence colocalization analysis. Blue: DAPI; red: fluorescent-labeled Albu-PTX or Lipo-PTX. Scale bar, 20 μm. C A series of Z-stack images showing the distributions of PTX nanoformulations in GC1 PDOs. Blue: DAPI; red: fluorescent-labeled Albu-PTX or Lipo-PTX. Scale bar, 30 μm

Fig. 6

In vivo antitumor effects of intratumorally injected PTX nanoformulations in the GC1 PDX model. A Schematic illustration of GC1 PDX construction and the experimental design used in evaluation of the PDX response to intratumoral injection of PTX nanoformulations. B Individual growth kinetics of PDX tumors in different treatment groups (n = 6). C Survival curves of the mice in B. D TUNEL assay results and IHC analysis of Ki-67 expression in different treatment groups. Scale bar, 200 μm. Data in (D) are presented as the means ± s.d. (n = 3), and P values are determined by log-rank (Mantel-Cox) test (C) or one-way ANOVA with Tukey post-hoc test (D). ***P < 0.001 and ****P < 0.0001