Next Article in Journal
Transcriptomic Analysis of Adult Mouse Cardiac Stromal Cells Using Single-Cell qRT-PCR
Previous Article in Journal
Brain Ketone Bodies in Health, Evolution and Disease
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Cells
Volume 15
Issue 4
10.3390/cells15040383
cells-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Comprehensive Adenoid Cystic Carcinoma-Derived Organoid Platform for Disease Modeling and Drug Screening Captures Interpatient Heterogeneity

by
Yingyue Chai
1,2,3,4,†,
Yi Sui
1,2,3,4,†,
Xinyuan Zhang
1,2,3,4,
Shang Xie
1,2,3,4,
Yifan Kang
1,2,3,4,
Yanrui Feng
2,3,4,5,
Xiaofeng Shan
1,2,3,4,* and
Zhigang Cai
1,2,3,4,*
1
Department of Oral and Maxillofacial Surgery and Oral Emergency, Peking University School and Hospital of Stomatology, Beijing 100081, China
2
National Center for Stomatology, Beijing 100081, China
3
National Clinical Research Center for Oral Diseases, Beijing 100081, China
4
National Engineering Research Center of Oral Biomaterials and Digital Medical Devices, Beijing 100081, China
5
Department of Central Laboratory, Peking University School and Hospital of Stomatology, Beijing 100081, China
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Cells 2026, 15(4), 383; https://doi.org/10.3390/cells15040383
Submission received: 4 January 2026 / Revised: 11 February 2026 / Accepted: 20 February 2026 / Published: 23 February 2026
(This article belongs to the Section Stem Cells)
Browse Figures
Versions Notes

Highlights

What are the main findings?
  • Successfully established a stable, high-success-rate, and high-throughput comprehensive platform for ACC organoids culture.
  • ACC organoids platform enables efficient screening of multiple drugs within two weeks.
What are the implications of the main findings?
  • Providing an in vitro research model for personalized precision medicine in ACC, contributing to a deeper understanding of tumor biological characteristics.
  • The rapid drug screening system holds promise as a clinical tool for medication guidance, providing a scientific basis for safer and more effective patient treatment.

Abstract

Salivary adenoid cystic carcinoma (ACC) is a highly aggressive salivary gland malignancy characterized by infiltrative growth patterns that hinder complete resection. Lacking effective chemotherapy, recurrent or metastatic ACC remains clinically incurable. This research aimed to develop an efficient culture system for ACC organoids, which can preserve tumor heterogeneity and establish a reliable drug screening platform. Under our optimized culture conditions, ACC organoids grew rapidly and successfully recapitulated three characteristic histopathological patterns. Whole-genome sequencing (WGS) further confirmed they mirrored the genomic features of their parental tumors, including significantly mutated genes, non-coding regulatory region mutations, copy number variation, and minor allele frequency. RNA sequencing confirmed that ACC organoids recapitulated the MYB-NFIB fusion gene. At the protein level, these organoids contained multiple cellular components, including epithelial cells, mesenchymal cells, K7+ duct cells, a-SMA+ myoepithelial cells, K5+ basement membrane cells, and CD44+ tumor stem cells, with proper spatial distribution patterns. With an 88% success rate, the first ACC organoid platform, incorporating normal salivary gland (SG) organoids as toxicity controls, enabled high-throughput drug testing within two weeks. In conclusion, we developed an efficient culture system for ACC organoids that can preserve tumor heterogeneity and establish a reliable drug screening platform for mechanistic studies and personalized precision therapy research.

Graphical Abstract

1. Introduction

Salivary adenoid cystic carcinoma (ACC) is one of the most prevalent epithelial-derived malignant tumors of the salivary gland in the oral and maxillofacial region [1]. ACC usually originates from minor salivary glands, with the hard palate and nasal sinuses being the most common sites of occurrence. Among major salivary gland tumors, the parotid gland is most frequently affected [2]. Histopathologically, ACC is characterized by the proliferation of ductal and myoepithelial cells in tubular, cribriform, or solid patterns [3]. ACC shows an indolent progression, frequently presenting without symptoms in early stages. However, ACC possesses biological characteristics such as neurotropic invasion, easy local recurrence, and distant metastasis. The infiltrative growth pattern frequently precludes achieving clear surgical margins, leading to a poor long-term prognosis [4,5,6,7]. The 10-year survival rate for ACC patients is 52%, while the 15-year survival rate drops to 28% [8,9]. Despite recent advancements in tumor immunotherapy, targeted therapy, and neoadjuvant approaches, ACC still lacks effective chemotherapy drugs and comprehensive treatment programs [10,11,12,13].
Cancer research models play crucial roles across the entire translational spectrum, from fundamental investigations to drug discovery and clinical therapeutic strategies [14]. Tumor organoid models in 3D conditions, derived from a small amount of patient tissue, can personalize the tissue characteristics of different patients at any cancer development stage [15,16,17]. This method is simple, inexpensive, and widely applicable, offering high throughput and scalability, making it an excellent preclinical model [18,19,20]. Currently, for cancers such as liver [21,22], bladder [23,24], and gastric [25,26], numerous studies have shown that tumor sensitivity to drugs or treatments in vitro aligns closely with the patient’s clinical response to the corresponding treatment plan. The sensitivity to cytotoxic drugs is higher, effectively guiding the development of personalized precision medicine [27,28,29].
Existing ACC models primarily consist of a few cell line models and patient-derived xenotransplantation (PDX) models [30,31,32,33]. Regarding ACC organoids, Takada and colleagues harvested ACC samples from seven surgical patients, combining patient-derived tumor xenograft (PDX) technology to successfully establish a short-term PDX-derived organoid model of ACC, validating the potential for in vitro drug screening [34]. The research and development of ACC organoid models are still in their infancy, with low success rates and efficiency, and a lack of comprehensive verification at protein, RNA, and DNA levels [35], which restricts their application in screening many potential drugs. In addition, ACC organoid culture predominantly targets epithelial cells, yet it lacks the mesenchymal tissue cells in the tumor microenvironment, which are vital for cancer progression. These mesenchymal cells secrete soluble cytokines, extracellular matrix, and metabolites through intercellular interactions to regulate tumor cell proliferation, migration, invasion, and apoptosis [36,37]. A stable, efficient, and multicellular model construction and treatment prediction platform is needed for ACC research and patient care.
In response to the challenges outlined, our study aims to develop a simple system for culturing ACC organoids with a high success rate. This multicellular model system will faithfully recapitulate tumor architecture and pathological features while preserving tumor-specific characteristics and interpatient heterogeneity, thereby accurately mirroring in vivo responses. In conjunction with normal salivary gland (SG) organoids, we will establish a personalized compound ACC drug screening platform based on the ACC organoid model for high-throughput application and promotion. The successful implementation of this model will provide a crucial technical guarantee for ACC drug development and personalized treatment, enhance treatment efficacy, and foster significant advancements in ACC research, diagnosis, and treatment technology.

2. Materials and Methods

2.1. Human Specimens

Paired ACC and adjacent normal SG tissues were procured from surgical resections performed at the Department of Oral and Maxillofacial Surgery, Peking University School and Hospital of Stomatology (Beijing, China). Patients of all genders and ages who were definitively diagnosed with ACC via histopathological examination were included in the study, and those with a history of maxillofacial radiotherapy or chemotherapy were excluded from the study. This study did not involve randomization and blind methods. All patients provided informed consent prior to tissue acquisition. The institutional review board granted study permission before tissue collection. There was no attrition. Tissues were segmented into four parts for histology, RNA isolation, DNA isolation, and organoid culture.

2.2. Isolation and Culture of Human Salivary Gland Healthy and ACC Organoids

Healthy human SG and ACC tissues were collected during surgery and stored in DMEM (C11995500BT, Gibco, Grand Island, NY, USA) post-surgery. They were transferred to a biosafety cabinet, cut into small pieces (~0.5–1 mm3), and placed into centrifuge tubes with a digestion solution composed of 0.63 mg/mL Collagenase Type 2 (C8150, Solarbio, Beijing, China) and 0.5 mg/mL Hyaluronidase (37326-33-3, Solarbio, Beijing, China). Tissues were digested in a dry bath at 37 °C for 1 h, with the tubes oscillating every 10 min. Digestion was halted by centrifugation and washing with DMEM/F12 (C11330500BT, Gibco, Grand Island, NY, USA), after which the suspension was filtered through a 100 μm nylon cell strainer and centrifuged at 300–400 g for 5 min at 4 °C. A total of 10 µL of 0.4% Trypan Blue (T10282, Thermo Fisher Scientific, Grand Island, NY, USA) was mixed with 10 µL of cell suspension to calculate the number of live cells. Cells were then embedded in Matrigel (354234, Corning, Steuben, NY, USA) and seeded at a density of 7500–10,000 cells per 30 μL Matrigel per well in a low-attachment flat-bottom 96-well plate (3474, Corning, NY, USA). Tumor cells were divided into two parts: one was cultured in the specific ACC medium (ACCM) adapted from other human organoid culture systems [38], and the other was cultured in KM-S medium. The detailed composition of the culture medium is provided in Table 1. Healthy SG cells were also cultured in the ACCM at 37 °C in a 5% CO2 incubator.
The medium was changed every 3–4 days. Organoid spheres were passaged at a 1:3 split ratio with a time window of 14–21 days. To prepare passages, Cell Recovery Solution (354253, Gibco, Grand Island, NY, USA) was used to dissociate the organoids from the Matrigel for 40–60 min at 4 °C. Then, organoids were washed with Advanced DMEM/F12 and centrifuged at 300–400 g for 5 min at 4 °C, resuspended with Matrigel, and seeded as previously mentioned. Organoids were considered successfully established only after stable growth through more than three passages (P3). Upon reaching P4, they were cryopreserved for long-term storage. For frozen stock, the organoids were dissociated from the Matrigel and frozen with Organoid Cryopreservation Medium (E238023, bioGenous, Shanghai, China) in −70 °C. For organoid thawing, the cryovials were thawed in a 37 °C water bath until the cryopreservation medium was completely dissolved. The contents were then transferred to a centrifuge tube containing 5.0 mL of complete medium and centrifuged at 300 g for 5 min. After removing the supernatant, the pellet was resuspended in complete medium for cell counting. Then, the organoids were embedded in Matrigel at an appropriate density for subsequent culture. Images were captured using a Leica M80 stereoscope and a Leica MC170 HD camera.

2.3. Organoid Proliferation and Viability Assays

CellTiter-Glo (G756A, Promega, Madison, WI, USA) was used to assay the proliferation and viability of ACC organoids. The volume of CellTiter-Glo® 3D Reagent was equal to the volume of medium present in each well of the ACC organoids at days 4, 7, and 11. The contents were mixed for 5 min and incubated at room temperature for an additional 25 min. Each sample had three replicate wells, and the Varioskan LUX multimode reader (Thermo Fisher Scientific, Grand Island, NY, USA) was used for recording luminescence.

2.4. Hematoxylin and Eosin Staining

Tissues and organoids were fixed overnight at 4 °C with 4% (w/v) paraformaldehyde and sequentially dehydrated with 10%, 20%, and 30% sucrose, also overnight at 4 °C. Subsequently, the samples were frozen and sliced at a thickness of 5 μm and stored at −20 °C. Hematoxylin and eosin (HE) staining was performed on the samples following standard protocols.

2.5. Immunofluorescence and Confocal Microscopy

Slides of the aforementioned specimens were rehydrated with PBS for 30 min, then incubated in 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO, USA) for 7 min. Following this, the slices were washed three times with PBS, each wash lasting 3 min, and then blocked with 3% BSA for 2 h. Specifically, for panel 1 (Ep-Cam & Vimentin), a mixture was prepared using Anti-Ep-Cam (Rabbit anti-human, cat# ab223582, Abcam) and Anti-Vimentin (Mouse anti-human, cat# ab8069, Abcam). For panel 2 (K7 and a-SMA), a mixture was prepared using Anti-Cytokeratin 7 (Rabbit anti-human, cat# ab181598, Abcam) and Anti-α-SMA (Mouse anti-human, cat# ab7817, Abcam). For panel 3 (K5 and CD44), a mixture was prepared using Anti-Cytokeratin 5 (Rabbit anti-human, cat# ab128190, Abcam) and Anti-CD44 (Mouse anti-human, cat# ab6124, Abcam). For panel 4 (P63 and Ki67), a mixture was prepared using Anti-P63 (Rabbit anti-human, cat# ab124762, Abcam) and Anti-Ki67 (Mouse anti-human, cat# ab279653, Abcam). For panel 5 (MYB and P53), a mixture was prepared using Anti-c-MYB (Rabbit anti-human, cat# ab117635, Abcam) and Anti-P53 (Mouse anti-human, cat# ab1101, Abcam). For panel 6 (COLA1 and a-SMA), a mixture was prepared using Anti-Collagen 1 (Rabbit anti-human, cat# A22090, Abclonal) and Anti-α-SMA (Mouse anti-human, cat# ab7817, Abcam). For panel 7 (FN1 and VIM), a mixture was prepared using Anti-Fibronectin (Rabbit anti-human, cat# ab45688, Abcam) and Anti-Vimentin (Mouse anti-human, cat# ab8069, Abcam). For panel 8 (K5 and CD133), a mixture was prepared using Anti-Cytokeratin 5 (Rabbit anti-human, cat# ab128190, Abcam) and Anti-CD133 (Mouse anti-human, cat# ab264538, Abcam). For panel 9 (AQP5 and K19), a mixture was prepared using Anti-Aquaporin 5 (Rabbit anti-human, cat# ab78486, Abcam) and Anti-Cytokeratin 19 (Mouse anti-human, cat# ab7754, Abcam). For all panels, each pair of primary antibodies was diluted at a ratio of 1:200. The dilution buffer was prepared with a 1:1 mixture of 3% BSA and 0.1% Triton X-100. The samples were then washed with PBS as previously described. For ACC tissues and ACC organoids (panels 1–7), secondary antibodies (Alexa Fluor® 488-labeled goat anti-rabbit IgG (ZF-0511, ZSGB-BIO) and Alexa Fluor® 594-labeled goat anti-mouse IgG (ZF-0513, ZSGB-BIO)) were used. For SG tissues and SG organoids (panels 1, 2, 8, 9), Alexa Fluor® 594-labeled goat anti-rabbit IgG (ZF-0516, ZSGB-BIO) and Alexa Fluor® 488-labeled goat anti-mouse IgG (ZF-0512, ZSGB-BIO) were used. All secondary antibodies were diluted at a ratio of 1:100 in PBS and incubated for 1 h. Nuclei were then stained with 4′,6-diamidino-2-phenylindole (DAPI; C0065, Sigma Aldrich, St. Louis, MO, USA) for 5 min. All these procedures were conducted at room temperature. Images were captured using a laser scanning confocal microscope (Nikon, Tokyo, Japan).

2.6. Short Tandem Repeat (STR) Sequencing

DNA was extracted from ACC tissue and organoid samples using the Microread Genomic DNA Kit. The extracted DNA underwent PCR amplification of 20 STR loci and gender identification loci using the MicroreaderTM21 ID System. The PCR products were analyzed with an ABI 3730 × l Genetic Analyzer, and the results were interpreted using GeneMapperID-X software v1.6 (Applied Biosystems, Waltham, Massachusetts).

2.7. Whole-Genome Sequencing (WGS)

After the DNA library construction was completed, the insert size of the library was detected by the Agilent 5400 system (AATI), and after the insert size met expectations, the effective concentration of the library (1.5 nM) was accurately quantified by qPCR to ensure the quality of the library. When the library was qualified, the library was then sequenced on the Illumina Novaseq platform for PE150 according to the validated concentration and data output of the library. PE150 is Pair-end 150 bp, high-throughput sequencing. In the constructed DNA small fragment library, each insert fragment was sequenced at both ends, and each end was sequenced at 150 bp. Quality controls for data processing were as follows: (1) Discard a paired read if either read contains adapter contamination (>10 nucleotides aligned to the adapter, allowing ≤ 10% mismatches); (2) Discard a paired read if more than 10% of bases are uncertain in eitherone read; (3) Discard a paired read if the proportion of low quality (Phred quality < 5) bases is over 50% in either one read. Valid sequencing data were compared to the reference genome by BWA software v0.7.17(parameter: mem-t 4-k 32-M), and the comparison results were removed from duplicates by SAMTOOLS (parameter: rmdup). The sequencing achieved a depth of 30X and a coverage of over 98%. SAMTOOLS (-C 50-mpileup-m 2-F 0.002-d 1000) was used for individual SNP detection. In order to reduce the error rate of SNP detection, the following criteria were chosen for filtering: (1) the number of reads supported by SNPs was not less than 4; (2) the quality value (MQ) of SNPs was not less than 20. SAMTOOLs (mpileup-m 2-F 0.002-d 1000) detected small fragment insertions and deletions (InDel) less than 50 bp in length. CNV (copy number variation) was detected by CNVnator (parameter: -call 100). Insertions, deletions, inversions and translocations between samples and reference genomes can be detected by using BreakDancer software v1.3.6, based on the paired-end reads comparison to the relationship above the reference genome and the actual Insert size. The following are some of the main features of PE read insertion (INS), deletion (DEL), inversion (INV), intra-chromosomal translocation (ITX), and inter-chromosomal translocation (CTX). Filtering to remove SV results with less than 2 support for PE reads. ANNOVAR was used for functional annotation of variants.

2.8. RNA Sequencing

The library was checked with Qubit and real-time PCR for quantification and a bioanalyzer for size distribution detection. After library quality control, different libraries were pooled based on their effective concentration and targeted data amount, then subjected to Illumina sequencing. Raw data (raw reads) of fastq format were first processed through fastp software (0.23.4). Clean data (clean reads) were obtained by removing reads containing adapters, reads containing ploy-N and low-quality reads from raw data. At the same time, the Q20, Q30 and GC content of the clean data were calculated. HISAT2 (2.2.1) was used to build the index of the reference genome and to align paired-end clean reads to the reference genome. StringTie (v2.2.3) software was used to predict novel transcripts. FeatureCounts (2.0.6) was used to count the number of reads mapped to each gene. FPKM was used to estimate gene expression levels. Differential expression analysis for two conditions/groups was performed using the DESeq2 R package (1.42.0). The threshold of significant differential expression: padj ≤ 0.05, |log2(foldchange)| ≥ 1. ClusterProfiler (4.8.1) was used to test the statistical enrichment of differential expression genes in KEGG pathways. The GATK (v4.1.1.0) software was used to perform SNP calling.

2.9. RNA Extraction, cDNA Synthesis, and Quantitative Reverse Transcriptase-Polymerase Chain Reaction

Total RNA was isolated from organoids using the TRIzol reagent (10296010CN, Thermo Fisher Scientific, Grand Island, NY, USA ) as per the manufacturer’s instructions. Following the determination of quality and concentration, mRNA was reverse transcribed into cDNA using a RevertAid™ FirstStrand cDNA Synthesis Kit (Thermo) according to the manufacturer’s protocol. Quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) was performed with FastStart Universal SYBR Green Master (ROX) Reagent (Roche). The cycling conditions involved 95 °C for 10 min, followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s. Expression levels were calculated using the comparative cycle threshold (Ct) method, with the expression level of 28s serving as an internal control. The primers we used were as follows: forward primer: CAGACCCATTTACTTGTGTTGGA (MKI67), CCTCCTAGCCTTCGTTGGTG (MYB), CAGACTATGCCTGCAGCTGTG (NOTCH1), CGACCCCCTGTGAAGTGATT (JAG1), GGCCCTCATGGTATCTGCAC (NOTCH3), TTACAACACCCGAGCAAGGA (CMYC), CTTGCCCAGGAAGAGACAGG (TP63), CTGTGTCCTCAAGTGCTCCC (NRARP) CCCAGTGCTCTGAATGTCAA (28s) and reverse primer: ACGCCTGGTTACTATCAAAAGG (MKI67), AAAGCGCCTCGCCAGCAAG (MYB), CTGGCACGATTTCCCTGACC (NOTCH1), ACTCTTGCACTTCCCGTGAG (JAG1), CAGGCATGGGTTGGGGTC (NOTCH3), CGCGGGAGGCTGGTTT (CMYC), CGCTTCGTACCATCACCGTT (TP63), TCCTGCGTCACTTTCTGTCC (NRARP), AGTGGGAATCTCGTTCATCC (28s).

2.10. Drug Screening

Passage 1 (P1) organoids were used for drug screening. Organoids were resuspended with Matrigel and seeded into low-attachment U-shaped-bottom 96-well plates (Corning) at a concentration of 33,000 cells/mL per well. On day 5–7, once organoids had reached 200–300 µm in diameter, the medium of spheres was replaced with ACCM supplemented with Gemcitabine (HY-17026, MedChemExpress, South Brunswick, NJ, USA), Cyclophosphamide (S1217, Selleckchem, Houston, TX, USA), Cisplatin (S1166, Selleckchem, Houston, TX, USA), Doxorubicin (HY-15142, MedChemExpress, NJ, USA), Tamoxifen (HY-13757A, MedChemExpress, NJ, USA), 5-Fluorouracil (HY-90006, MedChemExpress, NJ, USA), Vincristine (HY-N0488A, MedChemExpress, NJ, USA), Paclitaxel (HY-B0015, MedChemExpress, NJ, USA), Docetaxel (S1148, Selleckchem, Houston, TX, USA), Regorafenib (HY-10331, MedChemExpress, NJ, USA), Lapatinib (S2111, Selleckchem, Houston, TX, USA), Lenvatinib (HY-10981, MedChemExpress, NJ, USA), and 10 µM of Staurosporine (HY-15141, MedChemExpress, NJ, USA) as a positive control, with ACCM or 0.1% DMSO (HY-Y0320, MedChemExpress, NJ, USA) serving as a negative control. After 6 days of drug incubation, ATP levels were quantified using the CellTiter-Glo® 3D Cell Viability Assay (G9683, Promega, Beijing, China) following the manufacturer’s instructions. Screens were conducted in both technical (same screening run) and biological duplicates. Luminescence readings were acquired using a multimode microplate reader. All screening plates underwent stringent quality-control measures, the experiments included at least three independent technical replicates, and the outliers were removed prior to statistical analysis. Cell viability data were normalized against the untreated negative control group (set as 100%), and a Z-factor score was calculated by comparing negative and positive-control wells. Dose–response curves were fitted using the four-parameter logistic (4PL) model by GraphPad Prism software (Version 8).

2.11. Statistical Analysis

All statistical analyses were conducted using SPSS software (version 29.0.1.0 (171)). A paired T-test for two independent samples (organoids derived from ACC and healthy SG tissues) was performed for qPCR statistical analysis. Chi-square test and Fisher’s exact test were employed to analyze factors influencing the success rate of organoid culture. A p-value of <0.05 indicated statistical significance. Results are presented as the means ± standard deviation (SD).

2.12. Ethics Approval and Consent to Participate

Ethics approval and consent to participate: Human tissues experiments were in volved in this study, and all the procedures were reviewed and approved by the Biomedical Ethics Committee of Peking University Stomatological Hospital in China on 10 January 2023 (permit Number: PKUSSIRB-202282160; title of approved project: Patient-Derived Organoid Models for Oral Adenocarcinoma: Development and Application in Personalized Drug Screening).

2.13. Generative AI Statement

Generative AI and AI-assisted technologies were NOT used in the preparation of this work.

3. Results

3.1. Construction and Optimization of the Organoid Culture System for ACC

Model construction for ACC, a slow-growing cancer, presents challenges, as its primary cells either grow slowly or die in vitro, and the ACC tissue is tough and hard to digest. Preserving cell viability post-digestion is crucial for organoid formation. Based on reports on ACC’s extracellular matrix components, we chose relatively mild hyaluronidase and collagenase II as the main constituents of the tissue digestive fluid, maintaining digestion time at about 1 h to ensure cell digestion and protect cell activity. Single cells were embedded in Matrigel and cultured with medium. Organoids capable of stable expansion beyond three passages were defined as successfully established. Using this method, 36% of tissue samples had a percentage of viable cells post-digestion between 10% and 50% with an 88.89% success rate for organoid culture, and 52% had a percentage of viable cells post-digestion over 50% with a 96.15% success rate. In total, 56% of tissue samples were digested into between 1 × 105 and 1 × 106 living cells with an 89.29% success rate, and 34% of tissue samples were digested into more than 1 × 106 living cells with a 100% success rate (Figure S1, Table 2). Both the viable cell rate (less than 10%) and the number of viable cells (less than 1 × 105) significantly impacted the success rate of organoid culture (Table S1). For ACC6, 8, 9, 16, 25, 41, no organoid formation was observed even on day 21. Other organoids were maintained up to P4 in vitro and then cryopreserved. Organoids culture durations before banking are detailed in Figure S2.
ACCM with Y-27632, typically added to enhance stem cell survival of stem cells [39], was utilized to culture organoids derived from the ACC-PDX model [34], and KM-S supported SG organoid growth in our previous study [40]. Both media types were anticipated to culture ACC organoids and were detected. Y-27632 was excluded to simplify the culture process and prevent cell differentiation or stem cell purification with inappropriate use in our study. As depicted in Figure 1A, primary cells from either solid-type or tubular-cribriform-type ACC could proliferate and form spheres in both media types in 3D conditions without Y-27632. Notably, the ACC spheres in ACCM exhibited a smoother and neater edge (Figure 1A) and demonstrated superior vitality and growth compared to those in KM-S, which was statistically significant (Figure 1B).
HE staining revealed that ACC spheres cultured in KM-S were loosely formed, lacking tumor-like structures (Figure 1C). In contrast, those in ACCM formed three structures: solid, tubule, and cribriform, closely mirroring the pathological manifestations of ACC (Figure 1D–F). Ep-Cam, typically used as an epithelial cell marker, a myoepithelial cell marker a-SMA, and Vimentin (VIM), Collagen 1 (COLA1), and Fibronectin (FN1) as mesenchymal cell markers, were present in both ACC cell spheres cultured in the two media. Interestingly, tumor spheres in ACCM displayed a clear spatial distribution of epithelial and mesenchymal components, consistently containing both solid and cavity structures, aligning with the HE results (Figure 1E and Figure S3). In conclusion, we believe that ACCM is more suitable for constructing ACC organoids, and future studies will proceed with ACCM as the medium.
Utilizing the digestive process and ACCM, we achieved an 88% success rate in culturing 44 organoids from 50 surgical specimens in vitro (Table 3). As the organoid construction protocol stabilized and proficiency improved, the success rate rose from 75% (15 successful out of 20 total) in the initial stage to 96.67% (29 successful out of 30 total) in the later stage (Table S2). Factors such as patient age, gender, tumor location, pathological tumor classification, T stage, lymph invasion, distant metastasis, and perineural invasion (PNI) were collected but did not significantly impact the ACC organoid culture success rate (Table 3). Table S3 details the failed samples. We observed that cribriform-type ACC constituted most failed samples, but this was not statistically significant.
Additionally, ACCM can facilitate the formation and growth of normal SG organoids with a success rate approaching 100% (Figure S4A). These organoids contain spatially distributed epithelial and mesenchymal cells, expressing normal salivary gland-specific markers: lumen cell markers K7, K19, CD133, basal cell marker K5, and myoepithelial cell marker a-SMA. Mirroring the salivary glands, CD133 was expressed in the cell tip inside lumens, K5+ cells were located on the basal side of the lumen, and a-SMA+ cells were distributed around the K7+ lumen-like structure (Figure S4B). This indicates that SG organoids not only contained salivary gland-specific cells but also exhibited a polar and spatially specific distribution. The ACC organoid culture system can be employed for constructing SG organoids. The uniform cultural conditions ensure and facilitate personalized and comprehensive drug screening for the ACC.

3.2. Similarity Verification of ACC Organoids and Source Tumors

The STR results of tumors and their derived organoids from three patients indicated no third-rank gene phenomenon at each locus, signifying no cross-contamination of other human cells. The matching rate of ACC tumors and their derived organoids was 100% at the amelogenin locus and 20 STR detection loci, affirming their common origin (Figure 2A and Figure S5).
Cancer genomes typically undergo unique and unpredictable events like multiple point mutations, insertions and deletions, translocations, and fusions. WGS was utilized to acquire sequencing data of the entire genome with uniform coverage of effective data from patient-derived ACC tissues, corresponding organoids, and normal SG tissues for further exploration of cancer genetic mutations. Significantly mutated genes (SMG) take into account somatic SNV and InDel variants, retaining genes with a Q-value of <0.2. Figure 2B illustrates that the high-frequency mutant genes of ACC organoids closely resemble those of the source tumor tissue. Somatic mutations in non-coding regulatory regions, which can activate protooncogenes or inactivate tumor suppressor genes by altering gene transcription levels, are also key mechanisms of tumor formation. Figure 2C demonstrates that ACC organoids almost perfectly inherit high-frequency mutated genes in non-coding regions from the source tissue. Additionally, somatic CNV, a significant somatic mutation in tumor samples, potentially promoting or inhibiting cell growth, proliferation, metastasis, and recurrence, was analyzed, as well as the MAF at SNP sites. The CNV and MAF of ACC tumors were essentially conserved in the derived organoids, capturing the gene heterogeneity that varies from patient to patient (Figure 2D,E). Moreover, ACC molecular biological studies have demonstrated that aberrant overexpression of the MYB gene and MYB-NFIB gene translocation are one of the characteristic features of ACC [41]. A WGS study of 30 Chinese ACC cases revealed that the MYB-NFIB fusion was identified in 8 out of 30 cases (26.7%), accounting for 21.6% of all detected structural variants [42]. Therefore, we analyzed the fusion genes in both the source tissues and matched organoids in our WGS data. Unfortunately, neither the primary tumors nor the corresponding ACC organoids from three patients (ACC22, ACC30, ACC68) detected the MYB-NFIB fusion gene through WGS (Figure S6). Nevertheless, these findings still demonstrate that cultured organoids can faithfully recapitulate the gene expression profiles of their parental tumor tissues.
RNA-seq was employed to further validate ACC organoids at high throughput levels. Principal component analysis results and correlation heat maps demonstrated the similarity of cultured ACC organoids to ACC tissues (ACC4, ACC10, ACC22) (Figure 2F,G). Notably, fusion gene prediction indicated that when the source ACC tumor (ACC4 and ACC10) itself had a specific MYB-NFIB fusion gene, organoids inherited this key feature (Figure 2H); meanwhile, the organoid culture process also introduced some gene fusions that did not present in the source tissue. Compared with normal tissues, we discovered that the KEGG of ACC organoids could individualize most of the key or specific differential gene enrichment signaling pathways from tumor tissues, such as micro RNAs in cancer, DNA replication, the p53 signaling pathway, the PI3K-Akt signaling pathway, natural killer cell-mediated cytotoxicity, and so on (Figure 2I).
Previous ChIP-seq analyses revealed that in ACC tissues, 81% of TP63 binding sites colocalize with MYB. Further bioinformatic analysis of potential target genes near MYB binding sites showed significant enrichment of NOTCH1 and its ligands JAG1 and JAG2 among MYB-associated targets [43]. These findings suggest that MYB, TP63, and the NOTCH1 signaling pathway may cooperatively regulate the transcriptomic profile of ACC. Additionally, the cell proliferation marker KI67 is highly expressed in 30–70% of ACC cases [44]. Through standardized processing of RNA-seq data and differential expression analysis, we demonstrated that the core components of the NOTCH signaling pathway (including NOTCH1, NOTCH3, JAG1, JAG2, HEY1, HEY2, MYC, NRARP), MYB signaling pathway (including MYB, BCL2), TP63 signaling pathway (including TP63, CDK6, TPRG1) and proliferation associated genes (including TERT, MKI67 and CD44) in primary ACC tumor tissues, which maintained persistently elevated expression patterns (p < 0.05) in matched patient-derived organoid (PDO) models (Figure 3A). This finding provides transcriptomic-level evidence that ACC organoid models stably preserve the critical molecular phenotypes of primary tumors. Comparative qRT-PCR analysis of KI67, MYB, TP63, and NOTCH1 pathway-related genes in ACC tumor tissues versus matched submandibular gland tissues, as well as in corresponding organoid models, demonstrated that ACC-derived organoids faithfully recapitulate the gene expression patterns of primary tumors (Figure 3B,C).
Next, we assessed the protein level similarity between ACC organoids and source tumors by examining cell composition and specific marker expression. ACC tumor cells originate from ductal and myoepithelial cells [3]. As depicted in Figure 3D, ACC organoids contain both K7+ ductal cells and a-SMA+ myoepithelial cells, mirroring ACC tumor tissues with spatial expression. K5, an epithelial progenitor marker, and CD44, a tumor stem marker [45], were tested to evaluate stemness, showing similar positive expression in cultured organoids (Figure 3E). Additionally, P53 is expressed in 40–70% of ACC patient tumors [46]. Thus, MYB, P63, P53, and KI67 were summarized as relatively specific markers to analyze whether ACC expression in organoids matched that of the source tumor. Figure 3F,G present a case of ACC tumor with P63(+), KI67(+), MYB (+), and P53(−) expression, and the corresponding organoids exhibit the same expression. The negative control with secondary antibodies is shown in Figure S7.
In conclusion, we verified that the ACC cell spheres, formed by our construction protocol, were ACC organoids. These organoids closely mirrored the source tissues at the DNA, RNA, protein, and morphology levels and effectively captured interpatient heterogeneity.

3.3. Establishment of a Personalized, Comprehensive ACC Drug Screening Platform Based on Organoids

Initially, we selected four potentially effective drugs—regorafenib, gemcitabine, cisplatin, and vincristine—to analyze the IC50 values of ACC organoids from eight patients. Figure 4A illustrates that ACC from different patients exhibited individual drug responses. Specifically, ACC22 was sensitive to gemcitabine, ACC29 was most sensitive to regorafenib, ACC30 was most sensitive to vincristine, and ACC43 was most sensitive to gemcitabine. However, the IC50 values of ACC19 and ACC28 for all four drugs exceeded 1 µM, and those of ACC36 and ACC44 were even more than 10 µM (Figure 4B), indicating the need for a broader range of drug options. Cisplatin was ineffective for all the ACC organoids, with IC50 values greater than 1 µM. To establish a comprehensive drug screening platform, we introduced normal SG organoids as a control for drug toxicity. Unlike the individualized drug response of tumor organoids between patients, the IC50 values of SG organoids from different patients for the same drug were essentially consistent (Figure 4C,D). Using the IC50 value of SG organoids as a drug toxicity benchmark, Figure 4E visually demonstrated the sensitivity of ACC organoids to drugs in different patients. The Z-factor, indicating high throughput test quality, showed a value greater than 0.4, signifying high organoid drug screening quality. The organoids from both ACC and normal SG met the drug sensitivity test standards (Figure 4F).
Considering intratumoral heterogeneity, we sought tumors larger than 4 cm in diameter and conducted organoid culture and drug sensitivity tests at both the sampling center and periphery. This was to evaluate if this sampling method and organoids cultured at different tumor locations would influence drug sensitivity results. Using ACC organoids from different patients as mutual controls, drug susceptibility detection results varied significantly between patients. However, ACC organoids from different parts of the same tumor responded similarly to drugs, with IC50 values less than 10-fold apart (Figure 5A,B).
We further broadened the range of drugs to test our organoid platform’s detectability. Organoids derived from a single sample could simultaneously meet the drug sensitivity detection of 12 drugs and 5 gradients. After six days of treatment with varying drug concentrations starting on culture day 7, the appearance of ACC and normal SG organoids is shown in Figure 5C and Figure S5. Positive control (staurosporine 10 µM) and negative control (ACCM, 0.1% DMSO) are also depicted in Figure S8. Additionally, only one patient with separate palatal and maxillary sinus tumors had been collected in the multi-year study to test intertumoral heterogeneity, and was administered 12 drugs. However, the organoids derived from ACC at different sites in this patient showed a similar response to nine drugs, and the difference in IC50 values for three drugs was more than 10-fold (Figure 5D).

4. Discussion

In this study, we successfully established a stable and efficient system for constructing ACC organoids. These organoids can simulate the tissue structure, cellular component, protein expression, and pathological state of the tumor and even capture interpatient heterogeneity at RNA and whole-genome levels with a high success rate. The first constructed organoid platform, containing ACC organoids and normal SG organoids, offers uniform culture conditions and easy operation. It can be used for high-throughput drug sensitivity detection to reflect the individualized drug sensitivity of ACC and the toxicity of healthy SG between patients.
ACC is a very specific cancer that progresses slowly but grows by leaps with high invasiveness. Surgical resection is the best option, but it often results in a positive margin [45]. When ACC metastasizes or relapses, chemotherapy is urgently needed but ineffective with existing treatment attempts [10,11,12,13]. Organoid technology provides hope for overcoming the difficult culture of primary ACC cells in vitro and constructing a mature model of ACC. Unlike conventional organoid culture systems [47], we modified the tissue digestion protocol to maximize cell viability while maintaining sufficient cell yield, then we simplified the conventional medium composition, followed by rigorous medium screening to identify the optimal culture system for ACC organoids propagation. The success rate of ACC organoid culture was stable at 88%, and the success rate exceeded 95% in the later stage of the study. Compared to existing studies on ACC organoids [34,48], our organoid biobank exhibited significant improvements in both biobank scale (50 cases) and culture success rate (>85%). These advances provided an essential experimental basis for developing large-scale, high-throughput drug screening platforms, constituting a major innovation in ACC organoids research. It was observed that when the cell viability rate was less than 10%, and the number of living cells was less than 1 × 105, it did affect the success rate of organoid construction, and it was recommended to take measures in the process of organoid construction.
The comprehensive analysis, from morphological structure and multi-omics analysis to protein expression, validates the authenticity of the personalized ACC organoids constructed. We confirmed that the ACC organoids faithfully recapitulate the expression profiles of key regulatory factors in ACC pathogenesis [49], including NOTCH1, MYB-NFIB, and TP63, thereby validating the model’s ability to retain tumor-specific genetic alterations. Compared to normal SG tissues, the similarity of differential gene expression enrichment analysis between ACC organoids and source tumors is valuable for sensitive RNA expression and regulation. The ACC organoid bank can facilitate the study of tumor occurrence and development, alleviating the limitations of clinical samples. At the protein expression level, the ACC organoids contained epithelial cells, mesenchymal cells, K7+ duct cells, a-SMA+ myoepithelial cells, K5+ basement membrane cells, and CD44+ tumor stem cells with spatial distribution, retaining the characteristic marker expression patterns of the parental tissue.
We implemented the ACC organoid culture process for normal SG organoid cultures, achieving a success rate near 100%. Both ACC organoids and SG organoids were integrated to form a comprehensive drug screening platform for drug sensitivity and toxicity detection. Drug sensitivity results revealed significant differences among patients for tumor organoids, while SG organoids from different patients exhibited similar drug sensitivity. This allowed us to continually expand the patient sample size for personalized drug screening, using the limited toxicity test results of normal salivary gland organoids as a benchmark to alleviate funding and time costs.
However, the absence of validated chemotherapeutic agents remains a major barrier in ACC treatment. DODD et al. summarized the single drug response rate of ACC from high to low as fluoropyridine 39.5% (33–46%), cisplatin 35% (0–70%), anthracyclines 26. 5% (10–43%), and alkylating agents 0% [50]. In patients with recurrence and metastasis, regardless of single-drug chemotherapy or multi-drug combination, the objective response rate of the ACC treatment regimen was no more than 20% [51]. Twelve clinically common drugs tested in our organoid platform faced challenges identical to the therapeutic obstacles encountered in clinical treatment, including low drug sensitivity, severe toxicity, and side effects of semi-inhibitory concentration, necessitating the expansion of the drug library or the development of new drugs. Notably, cyclophosphamide—an alkylating agent—demonstrated no inhibitory activity against ACC organoids, thereby mirroring clinical observations showing its limited efficacy in ACC treatment. Currently, the ACC organoids drug screening platform we have established is capable of screening multiple drug combinations within two weeks, with great potential to expand to more targeted drugs, radiopharmaceuticals, and high-throughput small molecule screening. The emergence of the ACC organoids platform is expected to reduce costs, enhance clinical efficiency, decrease toxic side effects, and advance the research and development process.
There are also some limitations in this study. Firstly, although the ACC organoids contained multiple cells, we did not manage to detect endothelial cells and immune cells, indicating that introducing co-culture technologies, such as organoid chips, to create a more elaborate culture condition is necessary for reconstructing the tumor microenvironment [52,53]. In addition, ACC organoids from center/edge sampling did not display significant drug susceptibility heterogeneity within tumors. It might be the reason that the method of sampling and culture of organoids separately failed to simulate factors such as hypoxia niches, PH changes, and vascular distribution in the center or edge parts [54,55]. Hence, the drug sensitivity results reflected the comprehensive drug sensitivity of the entire tumor, so further models are needed to capture the intratumor heterogeneity. Furthermore, as ACC lacks a standard chemotherapy regimen, the ACC organoids-based drug screening platform cannot execute consistency evaluation with clinical data for the same patient, so the IC50 values obtained in this study are interpreted as comparative rather than predictive. Despite demonstrating technical feasibility, the clinical applicability of the ACC organoid-based drug screening platform remains challenged by the paucity of validated chemotherapeutic options. Therefore, further research should prioritize exploring drug combinations, developing tumor-targeted radiopharmaceuticals, and expanding drug category screening.

5. Conclusions

In summary, we introduce a stable, high success rate, and high-throughput organoid culture-based comprehensive platform for ACC. This platform enables a comprehensive investigation of subtype-specific signaling dependencies and guides personalized drug screening. Leveraging the platform’s high culture success rate and high-throughput screening capacity, future studies could employ longitudinal culturing to monitor subtype-dependent therapeutic responses and correlate organoid drug sensitivity profiles with patient outcomes. These advances will ultimately facilitate the transition of ACC treatment from empirical approaches to molecularly guided precision strategies.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cells15040383/s1. Figure S1: The viable cell rate and the number of viable cells after ACC digestion; Figure S2: Culture period of ACC organoids (Days); Figure S3: Immunofluorescence of the ACC tissues and organoids in the expression of the a-SMA marker, COLA1 marker VIM marker, and FN1 marker; Figure S4: Identification and characterization of normal SG organoids in ACCM; Figure S5: STR analysis confirms genetic identity between primary ACC Tumors and Patient-Derived Organoids; Figure S6: Fusion genes of three patients’ ACC tissues and derived organoids; Figure S7: Negative control of immunofluorescence with secondary antibodies; Figure S8: Effects of different drugs used in the drug screening on normal SG organoids; Table S1: Effect of the viable cell rate and the number of viable cells after tumor digestion on the success rate of ACC organoid culture; Table S2: The success rate of ACC organoid culture in different stages; Table S3: The clinical characteristics and tumor digestion of patients failed in ACC organoid culture.

Author Contributions

Conceptualization, Z.C., Y.S., S.X. and X.S.; Methodology, Y.C., Y.S. and X.Z.; Validation, Y.C., Y.S., Y.F. and Z.C.; Formal analysis, Y.C., Y.S. and Y.K.; Investigation, Y.C., Y.S., X.Z. and Y.K.; Resources, S.X., Y.K., Y.F. and X.S.; Data curation, Y.C. and Y.S.; Writing—original draft preparation, Y.C. and Y.S.; Writing—review and editing, Y.C., Y.S., S.X. and X.S.; Visualization, Y.C. and Y.S.; Supervision, Y.S., X. S. and Z.C.; Project administration, X.S.; Funding acquisition, Y.S. and Z.C. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Key R&D of Program of China (Grant NO. 2022YFC2504200) and the National Natural Science Foundation of China (No.82301023).

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Biomedical Ethics Committee of Peking University Stomatological Hospital in China (permit Number: PKUSSIRB-202282160; title of approved project: Patient-Derived Organoid Models for Oral Adenocarcinoma: Development and Application in Personalized Drug Screening; date of approval: 10 January 2023).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The RNA-seq datasets used and analyzed during the current study are available in GEO database at GSE319175.

Acknowledgments

We would like to thank Microread (info@microread.com) for STR authentication. We would like to thank Novogene (service@novogene.com) for RNA sequencing and WGS.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
ACCAdenoid cystic carcinoma
PDXPatient-derived xenotransplantation models
SGSalivary gland
ACCMACC Medium
KM-Skeratinocyte medium with supplements
HEHematoxylin and eosin
DAPI4′,6-diamidino-2-phenylindole
STRShort Tandem Repeat
WGSWhole-genome sequencing
CNVCopy number variation
MAFMinor allele frequency
qRT-PCRQuantitative reverse transcriptase-polymerase chain reaction
CtCycle threshold
SDStandard deviation
VIMVimentin
PNIPerineural invasion
SMGSignificantly mutated genes
PDOPatient-derived organoid

References

  1. Tian, Z.; Li, L.; Wang, L.; Hu, Y.; Li, J. Salivary gland neoplasms in oral and maxillofacial regions: A 23-year retrospective study of 6982 cases in an eastern Chinese population. Int. J. Oral Maxillofac. Surg. 2010, 39, 235–242. [Google Scholar] [CrossRef]
  2. Atallah, S.; Casiraghi, O.; Fakhry, N.; Wassef, M.; Uro-Coste, E.; Espitalier, F.; Sudaka, A.; Kaminsky, M.C.; Dakpe, S.; Digue, L.; et al. A prospective multicentre REFCOR study of 470 cases of head and neck Adenoid cystic carcinoma: Epidemiology and prognostic factors. Eur. J. Cancer 2020, 130, 241–249. [Google Scholar] [CrossRef]
  3. de Morais, E.F.; de Farias Morais, H.G.; de Almeida Freitas, R.; Coletta, R.D. Prognostic Significance of Histopathological Parameters for Salivary Gland Adenoid Cystic Carcinoma. Dent. J. 2023, 11, 262. [Google Scholar] [CrossRef]
  4. Gao, M.; Hao, Y.; Huang, M.X.; Ma, D.Q.; Luo, H.Y.; Gao, Y.; Peng, X.; Yu, G.Y. Clinicopathological study of distant metastases of salivary adenoid cystic carcinoma. Int. J. Oral. Maxillofac. Surg. 2013, 42, 923–928. [Google Scholar] [CrossRef]
  5. Shen, C.; Xu, T.; Huang, C.; Hu, C.; He, S. Treatment outcomes and prognostic features in adenoid cystic carcinoma originated from the head and neck. Oral. Oncol. 2012, 48, 445–449. [Google Scholar] [CrossRef]
  6. Amit, M.; Binenbaum, Y.; Trejo-Leider, L.; Sharma, K.; Ramer, N.; Ramer, I.; Agbetoba, A.; Miles, B.; Yang, X.; Lei, D.; et al. International collaborative validation of intraneural invasion as a prognostic marker in adenoid cystic carcinoma of the head and neck. Head Neck 2015, 37, 1038–1045. [Google Scholar] [CrossRef] [PubMed]
  7. Suton, P.; Luksic, I. Prognostic value of elective neck dissection in adenoid cystic carcinoma of head and neck: A meta-analysis: A call for randomized trials and international consensus. Int. J. Oral. Maxillofac. Surg. 2021, 50, 1403–1407. [Google Scholar] [CrossRef] [PubMed]
  8. Ellington, C.L.; Goodman, M.; Kono, S.A.; Grist, W.; Wadsworth, T.; Chen, A.Y.; Owonikoko, T.; Ramalingam, S.; Shin, D.M.; Khuri, F.R.; et al. Adenoid cystic carcinoma of the head and neck: Incidence and survival trends based on 1973–2007 Surveillance, Epidemiology, and End Results data. Cancer 2012, 118, 4444–4451. [Google Scholar] [CrossRef] [PubMed]
  9. Fang, Y.; Peng, Z.; Wang, Y.; Gao, K.; Liu, Y.; Fan, R.; Zhang, H.; Xie, Z.; Jiang, W. Current opinions on diagnosis and treatment of adenoid cystic carcinoma. Oral. Oncol. 2022, 130, 105945. [Google Scholar] [CrossRef]
  10. Laurie, S.A.; Ho, A.L.; Fury, M.G.; Sherman, E.; Pfister, D.G. Systemic therapy in the management of metastatic or locally recurrent adenoid cystic carcinoma of the salivary glands: A systematic review. Lancet Oncol. 2011, 12, 815–824. [Google Scholar] [CrossRef]
  11. Sahara, S.; Herzog, A.E.; Nör, J.E. Systemic therapies for salivary gland adenoid cystic carcinoma. Am. J. Cancer Res. 2021, 11, 4092–4110. [Google Scholar]
  12. Wagner, V.P.; Ferrarotto, R.; Vargas, P.A.; Martins, M.D.; Bingle, C.D.; Bingle, L. Drug-based therapy for advanced adenoid cystic carcinoma: Current landscape and challenges based on an overview of registered clinical trials. Crit. Rev. Oncol. Hematol. 2023, 181, 103886. [Google Scholar] [CrossRef] [PubMed]
  13. Dewenter, I.; Otto, S.; Kakoschke, T.K.; Smolka, W.; Obermeier, K.T. Recent Advances, Systemic Therapy, and Molecular Targets in Adenoid Cystic Carcinoma of the Head and Neck. J. Clin. Med. 2023, 12, 1463. [Google Scholar] [CrossRef]
  14. Shinozawa, T.; Kimura, M.; Cai, Y.; Saiki, N.; Yoneyama, Y.; Ouchi, R.; Koike, H.; Maezawa, M.; Zhang, R.R.; Dunn, A.; et al. High-Fidelity Drug-Induced Liver Injury Screen Using Human Pluripotent Stem Cell-Derived Organoids. Gastroenterology 2021, 160, 831–846.e10. [Google Scholar] [CrossRef]
  15. Lancaster, M.A.; Knoblich, J.A. Organogenesis in a dish: Modeling development and disease using organoid technologies. Science 2014, 345, 1247125. [Google Scholar] [CrossRef]
  16. Fatehullah, A.; Tan, S.H.; Barker, N. Organoids as an in vitro model of human development and disease. Nat. Cell Biol. 2016, 18, 246–254. [Google Scholar] [CrossRef]
  17. LeSavage, B.L.; Suhar, R.A.; Broguiere, N.; Lutolf, M.P.; Heilshorn, S.C. Next-generation cancer organoids. Nat. Mater. 2022, 21, 143–159. [Google Scholar] [CrossRef]
  18. Drost, J.; Clevers, H. Organoids in cancer research. Nat. Rev. Cancer 2018, 18, 407–418. [Google Scholar] [CrossRef] [PubMed]
  19. Bleijs, M.; van de Wetering, M.; Clevers, H.; Drost, J. Xenograft and organoid model systems in cancer research. EMBO J. 2019, 38, e101654. [Google Scholar] [CrossRef] [PubMed]
  20. Takahashi, Y.; Inoue, Y.; Sato, S.; Okabe, T.; Kojima, H.; Kiyono, H.; Shimizu, M.; Yamauchi, Y.; Sato, R. Drug cytotoxicity screening using human intestinal organoids propagated with extensive cost-reduction strategies. Sci. Rep. 2023, 13, 5407. [Google Scholar] [CrossRef]
  21. Broutier, L.; Mastrogiovanni, G.; Verstegen, M.M.; Francies, H.E.; Gavarró, L.M.; Bradshaw, C.R.; Allen, G.E.; Arnes-Benito, R.; Sidorova, O.; Gaspersz, M.P.; et al. Human primary liver cancer-derived organoid cultures for disease modeling and drug screening. Nat. Med. 2017, 23, 1424–1435. [Google Scholar] [CrossRef] [PubMed]
  22. Zhang, C.J.; Meyer, S.R.; O’Meara, M.J.; Huang, S.; Capeling, M.M.; Ferrer-Torres, D.; Childs, C.J.; Spence, J.R.; Fontana, R.J.; Sexton, J.Z. A human liver organoid screening platform for DILI risk prediction. J. Hepatol. 2023, 78, 998–1006. [Google Scholar] [CrossRef] [PubMed]
  23. Lee, S.H.; Hu, W.; Matulay, J.T.; Silva, M.V.; Owczarek, T.B.; Kim, K.; Chua, C.W.; Barlow, L.J.; Kandoth, C.; Williams, A.B.; et al. Tumor Evolution and Drug Response in Patient-Derived Organoid Models of Bladder Cancer. Cell 2018, 173, 515–528.e17. [Google Scholar] [CrossRef] [PubMed]
  24. Medle, B.; Sjödahl, G.; Eriksson, P.; Liedberg, F.; Höglund, M.; Bernardo, C. Patient-Derived Bladder Cancer Organoid Models in Tumor Biology and Drug Testing: A Systematic Review. Cancers 2022, 14, 2062. [Google Scholar] [CrossRef]
  25. Yan, H.H.N.; Siu, H.C.; Law, S.; Ho, S.L.; Yue, S.S.K.; Tsui, W.Y.; Chan, D.; Chan, A.S.; Ma, S.; Lam, K.O.; et al. A Comprehensive Human Gastric Cancer Organoid Biobank Captures Tumor Subtype Heterogeneity and Enables Therapeutic Screening. Cell Stem Cell 2018, 23, 882–897.e11. [Google Scholar] [CrossRef]
  26. Zu, M.; Hao, X.; Ning, J.; Zhou, X.; Gong, Y.; Lang, Y.; Xu, W.; Zhang, J.; Ding, S. Patient-derived organoid culture of gastric cancer for disease modeling and drug sensitivity testing. Biomed. Pharmacother. 2023, 163, 114751. [Google Scholar] [CrossRef]
  27. Xu, H.; Jiao, D.; Liu, A.; Wu, K. Tumor organoids: Applications in cancer modeling and potentials in precision medicine. J. Hematol. Oncol. 2022, 15, 58. [Google Scholar] [CrossRef]
  28. Yokoi, R.; Shibata, M.; Odawara, A.; Ishibashi, Y.; Nagafuku, N.; Matsuda, N.; Suzuki, I. Analysis of signal components <500 Hz in brain organoids coupled to microelectrode arrays: A reliable test-bed for preclinical seizure liability assessment of drugs and screening of antiepileptic drugs. Biochem. Biophys. Rep. 2021, 28, 101148. [Google Scholar]
  29. Rassomakhina, N.V.; Ryazanova, A.Y.; Likhov, A.R.; Bruskin, S.A.; Maloshenok, L.G.; Zherdeva, V.V. Tumor Organoids: The Era of Personalized Medicine. Biochemistry 2024, 89, S127–S147. [Google Scholar] [CrossRef]
  30. Warner, K.A.; Oklejas, A.E.; Pearson, A.T.; Zhang, Z.; Wu, W.; Divi, V.; Rodriguez-Ramirez, C.; Castilho, R.M.; Polverini, P.J.; Nör, J.E. UM-HACC-2A: MYB-NFIB fusion-positive human adenoid cystic carcinoma cell line. Oral Oncol. 2018, 87, 21–28. [Google Scholar] [CrossRef]
  31. Li, J.; Perlaky, L.; Rao, P.; Weber, R.S.; El-Naggar, A.K. Development and characterization of salivary adenoid cystic carcinoma cell line. Oral Oncol. 2014, 50, 991–999. [Google Scholar] [CrossRef]
  32. Sahara, S.; Warner, K.A.; Herzog, A.E.; Zhang, Z.; Nör, J.E. Therapeutic inhibition of Bmi-1 ablates chemoresistant cancer stem cells in adenoid cystic carcinoma. Oral Oncol. 2023, 142, 106437. [Google Scholar] [CrossRef]
  33. Guimarães, L.D.; Webber, L.P.; Gaio, E.J.; Junior, D.S.; Gonçalves, P.; Wick, M.J.; Burr, N.S.; Squarize, C.H.; Castilho, R.M. Using PDX animal models to identify and stratify adenoid cystic carcinoma patients presenting an enhanced response to HDAC inhibitors. Am. J. Cancer Res. 2023, 13, 143–160. [Google Scholar]
  34. Takada, K.; Aizawa, Y.; Sano, D.; Okuda, R.; Sekine, K.; Ueno, Y.; Yamanaka, S.; Aoyama, J.; Sato, K.; Kuwahara, T.; et al. Establishment of PDX-derived salivary adenoid cystic carcinoma cell lines using organoid culture method. Int. J. Cancer 2021, 148, 193–202. [Google Scholar] [CrossRef]
  35. Lassche, G.; van Boxtel, W.; Aalders, T.W.; van Hooij, O.; van Engen-van Grunsven, A.C.H.; Verhaegh, G.W.; van Herpen, C.M.L.; Schalken, J.A. Development and characterization of patient-derived salivary gland cancer organoid cultures. Oral Oncol. 2022, 135, 106186. [Google Scholar] [CrossRef]
  36. Al-Raimi, H.A.I.; Kong, J.; Ran, Y.; Zhu, L.; Li, J.; Liu, X.; Yang, X.; Qi, D.; Liu, T. Extracellular Vesicles from Carcinoma-associated Fibroblasts Promote EMT of Salivary Adenoid Cystic Carcinoma Via IL-6. Arch. Med. Res. 2023, 54, 27–36. [Google Scholar] [CrossRef]
  37. Goto, H.; Nishioka, Y. Fibrocytes: A Novel Stromal Cells to Regulate Resistance to Anti-Angiogenic Therapy and Cancer Progression. Int. J. Mol. Sci. 2017, 19, 98. [Google Scholar] [CrossRef]
  38. Kim, J.; Koo, B.K.; Knoblich, J.A. Human organoids: Model systems for human biology and medicine. Nat. Rev. Mol. Cell Biol. 2020, 21, 571–584. [Google Scholar] [CrossRef] [PubMed]
  39. Wu, Y.; Shu, J.; He, C.; Li, M.; Wang, Y.; Ou, W.; He, Y. ROCK inhibitor Y27632 promotes proliferation and diminishes apoptosis of marmoset induced pluripotent stem cells by suppressing expression and activity of caspase 3. Theriogenology 2016, 85, 302–314. [Google Scholar] [CrossRef] [PubMed]
  40. Sui, Y.; Zhang, S.; Li, Y.; Zhang, X.; Hu, W.; Feng, Y.; Xiong, J.; Zhang, Y.; Wei, S. Generation of functional salivary gland tissue from human submandibular gland stem/progenitor cells. Stem Cell Res. Ther. 2020, 11, 127. [Google Scholar] [CrossRef] [PubMed]
  41. Ho, A.S.; Kannan, K.; Roy, D.M.; Morris, L.G.; Ganly, I.; Katabi, N.; Ramaswami, D.; Walsh, L.A.; Eng, S.; Huse, J.T.; et al. The mutational landscape of adenoid cystic carcinoma. Nat. Genet. 2013, 45, 791–798. [Google Scholar] [CrossRef]
  42. Wang, S.; Yu, Y.; Fang, Y.; Huang, H.; Wu, D.; Fang, H.; Bai, Y.; Sun, C.; Yu, A.; Fan, Q.; et al. Whole-exome sequencing reveals genetic underpinnings of salivary adenoid cystic carcinoma in the Chinese population. J. Genet. Genom. 2020, 47, 397–401. [Google Scholar] [CrossRef]
  43. Drier, Y.; Cotton, M.J.; Williamson, K.E.; Gillespie, S.M.; Ryan, R.J.; Kluk, M.J.; Carey, C.D.; Rodig, S.J.; Sholl, L.M.; Afrogheh, A.H.; et al. An oncogenic MYB feedback loop drives alternate cell fates in adenoid cystic carcinoma. Nat. Genet. 2016, 48, 265–272. [Google Scholar] [CrossRef]
  44. Seethala, R.R.; Hunt, J.L.; Baloch, Z.W.; Livolsi, V.A.; Leon Barnes, E. Adenoid cystic carcinoma with high-grade transformation: A report of 11 cases and a review of the literature. Am. J. Surg. Pathol. 2007, 31, 1683–1694. [Google Scholar] [CrossRef]
  45. Saleh, E.; Ukwas, A. Adenoid Cystic Carcinoma of Salivary Glands: A Ten-Year Review and an Assessment of the Current Management, Surgery, Radiotherapy, and Chemotherapy. Int. J. Otolaryngol. 2023, 2023, 7401458. [Google Scholar] [CrossRef] [PubMed]
  46. Liu, H.X.; Wu, M.; Sun, Y.M.; Han, F.Y.; Liu, Y.F.; Zhang, G.Z. Prognostic value of human papillomavirus infection and p53, p16, epidermal growth factor receptor and p34cdc2 expression in patients with salivary adenoid cystic carcinoma. Int. J. Clin. Exp. Pathol. 2017, 10, 7882–7889. [Google Scholar]
  47. Driehuis, E.; Spelier, S.; Beltrán Hernández, I.; de Bree, R.; Willems, S.M.; Clevers, H.; Oliveira, S. Patient-Derived Head and Neck Cancer Organoids Recapitulate EGFR Expression Levels of Respective Tissues and Are Responsive to EGFR-Targeted Photodynamic Therapy. J. Clin. Med. 2019, 8, 1880. [Google Scholar] [CrossRef]
  48. Aizawa, Y.; Takada, K.; Aoyama, J.; Sano, D.; Yamanaka, S.; Seki, M.; Kuze, Y.; Ramilowski, J.A.; Okuda, R.; Ueno, Y.; et al. Establishment of experimental salivary gland cancer models using organoid culture and patient-derived xenografting. Cell Oncol. 2023, 46, 409–421. [Google Scholar] [CrossRef]
  49. Ho, A.S.; Ochoa, A.; Jayakumaran, G.; Zehir, A.; Valero Mayor, C.; Tepe, J.; Makarov, V.; Dalin, M.G.; He, J.; Bailey, M.; et al. Genetic hallmarks of recurrent/metastatic adenoid cystic carcinoma. J. Clin. Invest. 2019, 129, 4276–4289. [Google Scholar] [CrossRef] [PubMed]
  50. Dodd, R.L.; Slevin, N.J. Salivary gland adenoid cystic carcinoma: A review of chemotherapy and molecular therapies. Oral Oncol. 2006, 42, 759–769. [Google Scholar] [CrossRef] [PubMed]
  51. Bradley, P.J. Adenoid cystic carcinoma evaluation and management: Progress with optimism! Curr. Opin. Otolaryngol. Head Neck Surg. 2017, 25, 147–153. [Google Scholar] [CrossRef]
  52. Hofer, M.; Lutolf, M.P. Engineering organoids. Nat. Rev. Mater. 2021, 6, 402–420. [Google Scholar] [CrossRef] [PubMed]
  53. Jiang, X.; Ren, L.; Tebon, P.; Wang, C.; Zhou, X.; Qu, M.; Zhu, J.; Ling, H.; Zhang, S.; Xue, Y.; et al. Cancer-on-a-Chip for Modeling Immune Checkpoint Inhibitor and Tumor Interactions. Small 2021, 17, e2004282. [Google Scholar] [CrossRef] [PubMed]
  54. Wang, Z.; Zhang, S.; Kong, Z.; Li, S.; Sun, J.; Zheng, Y.; He, Z.; Ye, H.; Luo, C. Self-adaptive nanoassembly enabling turn-on hypoxia illumination and periphery/center closed-loop tumor eradication. Cell Rep. Med. 2023, 4, 101014. [Google Scholar] [CrossRef] [PubMed]
  55. Pernia Marin, M.; Salvatore, M. Analogies between the periphery of cancer and the leading edge of pulmonary fibrosis. J. Transl. Med. 2023, 21, 274. [Google Scholar] [CrossRef]
Cells 15 00383 g001
Figure 1. Construction and optimization of the organoid culture system for ACC. (A) Phase-contrast images showing the growth and phenotypes of ACC organoids in ACCM and KM-S. The top organoids were derived from the solid-type ACC, and the bottom organoids were derived from the tubular-cribriform-type ACC. Scale bar, 200 μm. (B) Organoid formation viability for ACC in ACCM and KM-S. The graph represents the mean ± SD of the luminescence. *** p < 0.001, * p < 0.05, n = 3. Scale bar, 200 μm. (C–E) Representative HE staining comparing ACC organoids in KM-S (C) and ACCM (D). Matching ACC tumors on three classical pathological types (E). (F) Immunofluorescence staining of organoids derived from the solid-type ACC and the tubular-cribriform-type ACC for the Ep-Cam epithelial marker and the VIM marker, and corresponding brightfield microscopy images (bottom). Scale bar, 200 μm.
Figure 1. Construction and optimization of the organoid culture system for ACC. (A) Phase-contrast images showing the growth and phenotypes of ACC organoids in ACCM and KM-S. The top organoids were derived from the solid-type ACC, and the bottom organoids were derived from the tubular-cribriform-type ACC. Scale bar, 200 μm. (B) Organoid formation viability for ACC in ACCM and KM-S. The graph represents the mean ± SD of the luminescence. *** p < 0.001, * p < 0.05, n = 3. Scale bar, 200 μm. (C–E) Representative HE staining comparing ACC organoids in KM-S (C) and ACCM (D). Matching ACC tumors on three classical pathological types (E). (F) Immunofluorescence staining of organoids derived from the solid-type ACC and the tubular-cribriform-type ACC for the Ep-Cam epithelial marker and the VIM marker, and corresponding brightfield microscopy images (bottom). Scale bar, 200 μm.
Cells 15 00383 g001
Cells 15 00383 g002
Figure 2. STR and high-throughput similarity analysis of ACC organoids and source tumors by RNA-seq and WGS. (A) STR depicting 100% matching of ACC tumors and their derived organoids at the amelogenin locus and 20 STR detection loci. T, tissue; O, organoid. (B) SMG, including somatic SNV and InDel variants of the patient’s tumor tissues and corresponding organoids. The top bar chart shows the mutation rate of each sample. The heat map below shows high-frequency mutation genes and mutation types. The bar chart on the left side of the heat map shows the proportion of samples with mutations in each gene, and the right side shows the −log10(p-value) of the genes with high frequency mutations. T, tissue; O, organoid. (C) Somatic mutations in non-coding regulatory regions of the patient’s ACC tissues and corresponding organoids. The vertical coordinates represent different mutation locations in the non-coding region, and the mutation frequency of the sample at this site is indicated by the shade of red. (D) CNV of the patient’s ACC tissues and corresponding organoids. The horizontal coordinate is chromosomes 1–22. Blue, DNA copy number gains; red, DNA copy number losses. (E) Detailed CNV in the genome (the top scatterplot of each sample), and MAF at SNP sites (the bottom scatterplot of each sample) of ACC tissue-organoid pairs. In the CNV distribution scatterplot, red indicates an increase in copy number, and blue indicates a decrease in copy number. In the following MAF distribution scatterplot, orange indicates uniform distribution of AB alleles, blue indicates preference of AB alleles, and loss of heterozygosity occurs when MAF is separated to 0,1. (F) Principal component analysis of ACC organoids (ACC_O), source tumors (ACC_T), and normal SG tissues (SG_T) from three patients, using their projections onto the first two principal components (PC1 and PC2). Each data point represents one sample. PC1 shows a strong correlation between ACC organoids and source tumors (opposed to normal SG tissues). (G) Correlation heat maps between samples of ACC organoids, source tumors and normal SGs. (H) Fusion gene prediction of ACC organoids, source tumors and normal SGs. All the fusion gene names are enlarged in the image. The specific MYB-NFIB fusion gene in ACC tumors was inherited by their derived organoids. (I) Heatmap recapitulating representative differential gene enrichment signaling pathways significantly upregulated (purple) and downregulated (green) in ACC organoids and source tumors by KEGG analysis. Padj shows multiple hypothesis tests’ corrected p-values.
Figure 2. STR and high-throughput similarity analysis of ACC organoids and source tumors by RNA-seq and WGS. (A) STR depicting 100% matching of ACC tumors and their derived organoids at the amelogenin locus and 20 STR detection loci. T, tissue; O, organoid. (B) SMG, including somatic SNV and InDel variants of the patient’s tumor tissues and corresponding organoids. The top bar chart shows the mutation rate of each sample. The heat map below shows high-frequency mutation genes and mutation types. The bar chart on the left side of the heat map shows the proportion of samples with mutations in each gene, and the right side shows the −log10(p-value) of the genes with high frequency mutations. T, tissue; O, organoid. (C) Somatic mutations in non-coding regulatory regions of the patient’s ACC tissues and corresponding organoids. The vertical coordinates represent different mutation locations in the non-coding region, and the mutation frequency of the sample at this site is indicated by the shade of red. (D) CNV of the patient’s ACC tissues and corresponding organoids. The horizontal coordinate is chromosomes 1–22. Blue, DNA copy number gains; red, DNA copy number losses. (E) Detailed CNV in the genome (the top scatterplot of each sample), and MAF at SNP sites (the bottom scatterplot of each sample) of ACC tissue-organoid pairs. In the CNV distribution scatterplot, red indicates an increase in copy number, and blue indicates a decrease in copy number. In the following MAF distribution scatterplot, orange indicates uniform distribution of AB alleles, blue indicates preference of AB alleles, and loss of heterozygosity occurs when MAF is separated to 0,1. (F) Principal component analysis of ACC organoids (ACC_O), source tumors (ACC_T), and normal SG tissues (SG_T) from three patients, using their projections onto the first two principal components (PC1 and PC2). Each data point represents one sample. PC1 shows a strong correlation between ACC organoids and source tumors (opposed to normal SG tissues). (G) Correlation heat maps between samples of ACC organoids, source tumors and normal SGs. (H) Fusion gene prediction of ACC organoids, source tumors and normal SGs. All the fusion gene names are enlarged in the image. The specific MYB-NFIB fusion gene in ACC tumors was inherited by their derived organoids. (I) Heatmap recapitulating representative differential gene enrichment signaling pathways significantly upregulated (purple) and downregulated (green) in ACC organoids and source tumors by KEGG analysis. Padj shows multiple hypothesis tests’ corrected p-values.
Cells 15 00383 g002
Cells 15 00383 g003
Figure 3. RNA and protein expression of ACC organoids and source tumors. (A) Heatmap demonstrates differentially expressed genes in ACC tissues versus SG tissues across NOTCH signaling pathway, MYB pathway, TP63 pathway, and proliferation-associated genes. ACC organoids recapitulated identical expression trends compared to SG organoids. DEG threshold: |log2FC| >1, padj < 0.05. (B-C) qRT-PCR showing that compared to normal SG tissues, the expression of characteristic genes significantly increased in ACC tissues (B), and their derived ACC organoids exhibited a similar expression trend (C). Error bars, SD. *** p < 0.001, ** p < 0.01, * p < 0.05, n = 3. (D–G) Immunofluorescence of ACC tumors (top) and derived organoids (bottom) in the expression of the K7 ductal marker and the a-SMA myoepithelial marker (D), the K5 progenitor marker and the CD44 tumor stem marker (E), and the ACC characteristic markers P63 and Ki67 (F), MYB and P53 (G). Scale bar, 200 μm.
Figure 3. RNA and protein expression of ACC organoids and source tumors. (A) Heatmap demonstrates differentially expressed genes in ACC tissues versus SG tissues across NOTCH signaling pathway, MYB pathway, TP63 pathway, and proliferation-associated genes. ACC organoids recapitulated identical expression trends compared to SG organoids. DEG threshold: |log2FC| >1, padj < 0.05. (B-C) qRT-PCR showing that compared to normal SG tissues, the expression of characteristic genes significantly increased in ACC tissues (B), and their derived ACC organoids exhibited a similar expression trend (C). Error bars, SD. *** p < 0.001, ** p < 0.01, * p < 0.05, n = 3. (D–G) Immunofluorescence of ACC tumors (top) and derived organoids (bottom) in the expression of the K7 ductal marker and the a-SMA myoepithelial marker (D), the K5 progenitor marker and the CD44 tumor stem marker (E), and the ACC characteristic markers P63 and Ki67 (F), MYB and P53 (G). Scale bar, 200 μm.
Cells 15 00383 g003
Cells 15 00383 g004
Figure 4. Establishment of a personalized, comprehensive ACC drug screening platform combining ACC organoids with normal SG organoids. (A,B) Dose–response curves (A) and IC50 values (B) after 6 days of treatment with regorafenib, gemcitabine, cisplatin, and vincristine on ACC organoids from eight patients, generated from the luminescence signal intensities. Error bars, SD. n = 3. (C,D) Dose–response curves (C) and IC50 values (D) after 6 days of treatment with regorafenib, gemcitabine, cisplatin, and vincristine on normal SG organoids (characterization in Figure S8) from four patients, generated from the luminescence signal intensities. Error bars, SD. n = 3. (E) Heat map indicating the individual sensitivity of ACC organoids to drugs compared with the drug toxicity of normal SG organoids. (F) Z-factor values of the drug-screening data showed greater than 0.4, highlighting that the organoid platform met the drug sensitivity test standards. Each data point represents one sample.
Figure 4. Establishment of a personalized, comprehensive ACC drug screening platform combining ACC organoids with normal SG organoids. (A,B) Dose–response curves (A) and IC50 values (B) after 6 days of treatment with regorafenib, gemcitabine, cisplatin, and vincristine on ACC organoids from eight patients, generated from the luminescence signal intensities. Error bars, SD. n = 3. (C,D) Dose–response curves (C) and IC50 values (D) after 6 days of treatment with regorafenib, gemcitabine, cisplatin, and vincristine on normal SG organoids (characterization in Figure S8) from four patients, generated from the luminescence signal intensities. Error bars, SD. n = 3. (E) Heat map indicating the individual sensitivity of ACC organoids to drugs compared with the drug toxicity of normal SG organoids. (F) Z-factor values of the drug-screening data showed greater than 0.4, highlighting that the organoid platform met the drug sensitivity test standards. Each data point represents one sample.
Cells 15 00383 g004
Cells 15 00383 g005
Figure 5. The ACC drug screening platform allows high-throughput detection. (A,B) Dose–response curves (A) and IC50 values (B) of organoids separately derived from the center and the periphery of ACC tumors with regorafenib, gemcitabine, cisplatin, and vincristine. Error bars, SD. n = 3. (C) Representative images showing effects of 12 drugs with 5 concentration gradients on the same patient’s ACC organoids (Positive control and negative control detailed in Figure S8). Scale bar, 500 μm. (D) Summary of the 12 drugs used in the drug screening, the drug types, the associated pathway, the nominal targets and the screen IC50 results of organoids derived from one patient’s ACC tumors at the palatine (ACC21-Pal.) and the maxillary sinus (ACC21-MS), and the normal SG tissues. IC50 values highlighted in red are the differences in the response of the inter-tumor-derived organoids of more than 10-fold.
Figure 5. The ACC drug screening platform allows high-throughput detection. (A,B) Dose–response curves (A) and IC50 values (B) of organoids separately derived from the center and the periphery of ACC tumors with regorafenib, gemcitabine, cisplatin, and vincristine. Error bars, SD. n = 3. (C) Representative images showing effects of 12 drugs with 5 concentration gradients on the same patient’s ACC organoids (Positive control and negative control detailed in Figure S8). Scale bar, 500 μm. (D) Summary of the 12 drugs used in the drug screening, the drug types, the associated pathway, the nominal targets and the screen IC50 results of organoids derived from one patient’s ACC tumors at the palatine (ACC21-Pal.) and the maxillary sinus (ACC21-MS), and the normal SG tissues. IC50 values highlighted in red are the differences in the response of the inter-tumor-derived organoids of more than 10-fold.
Cells 15 00383 g005
Table 1. Medium formulation for ACC organoids.
Table 1. Medium formulation for ACC organoids.
Reagent NameConcentration
ACCM
DMEM/F12
HEPES (15630-080, Gibco, Grand Island, NY, USA)1%
L-Glutamine (25030-081, Gibco, Grand Island, NY, USA)1%
B27 Supplement (17504044, Gibco, Grand Island, NY, USA)2%
N-Acetyl-L-cysteine (A9165, Sigma-Aldrich, St. Louis, MO, USA)1 mM
Human R-Spondin-1 (CX83, Novoprotein, Suzhou, China)0.1 μg/mL
Human Wnt3a (C18K, Novoprotein, Suzhou, China)500 ng/mL
Human EGF (C029, Novoprotein, Suzhou, China)50 ng/mL
Human FGF-10 (CR11, Novoprotein, Suzhou, China)100 ng/mL
Antibiotic-Antimycotic (15240062, Thermo Fisher Scientific, Grand Island, NY, USA)1%
Nicotinamide (N0636, Sigma-Aldrich, St. Louis, MO, USA)10 mM
A83-01 (909910-43-6, Sigma-Aldrich, St. Louis, MO, USA)0.1 μM
Human Noggin (CB89, Novoprotein, Suzhou, China)0.1 μg/m
Dexamethasone (D4902, Sigma-Aldrich, St. Louis, MO, USA)1 μM
KM-S
Keratinocyte medium (2101, ScienCell, San Diego, CA, USA)
Bovine Serum Albumin(A8010, Solarbio, Beijing, China)5 μg/mL
Human FGFb (C046, Novoprotein, Suzhou, China)5 ng/mL
Human EGF1 ng/mL
Insulin (I6634, Sigma-Aldrich, St. Louis, MO, USA)5 μg/mL
Transferrin (T8158, Sigma-Aldrich, St. Louis, MO, USA)5 μg/mL
Hydrocortisone (H0888, Sigma-Aldrich, St. Louis, MO, USA)0.5 μg/mL
Table 2. Effect of the viable cell rate and the number of viable cells after tumor digestion on the success rate of ACC organoid culture.
Table 2. Effect of the viable cell rate and the number of viable cells after tumor digestion on the success rate of ACC organoid culture.
Rate of Viable CellsNumber of Viable Cells (104)
≤10%10–50%>50%≤1010–100>100
Totals6182652817
Success3162522517
Rate50.00%88.89%96.15%25.00%88.00%100.00%
p-value* (0.018)* (0.007)
Table 3. The clinical characteristics of patients and the success rate of ACC organoid culture.
Table 3. The clinical characteristics of patients and the success rate of ACC organoid culture.
TotalsSuccessRatep-Value
Gender
Male232086.96%>0.9999
Female272488.89%
Age
<50191789.47%0.146
≥50312787.10%
Site
Parotid gland55100.00%0.096
Submandibular gland5360.00%
Sublingual gland8787.50%
Palate222195.45%
Bucca6466.67%
Lingua44100.00%
Classification
Solid131292.31%0.405
Cribriform141178.57%
Tubular4375.00%
Cribriform-tubular191894.74%
T stage
122100.00%0.879
211981.82%
310990.00%
4272488.89%
Lymph invasion
Positive6466.67%0.146
Negative444090.91%
Distant metastasis
Positive2150.00%0.228
Negative484389.58%
PNI
Positive201995.00%0.381
Negative302583.33%
Totals504488%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Chai, Y.; Sui, Y.; Zhang, X.; Xie, S.; Kang, Y.; Feng, Y.; Shan, X.; Cai, Z. A Comprehensive Adenoid Cystic Carcinoma-Derived Organoid Platform for Disease Modeling and Drug Screening Captures Interpatient Heterogeneity. Cells 2026, 15, 383. https://doi.org/10.3390/cells15040383

AMA Style

Chai Y, Sui Y, Zhang X, Xie S, Kang Y, Feng Y, Shan X, Cai Z. A Comprehensive Adenoid Cystic Carcinoma-Derived Organoid Platform for Disease Modeling and Drug Screening Captures Interpatient Heterogeneity. Cells. 2026; 15(4):383. https://doi.org/10.3390/cells15040383

Chicago/Turabian Style

Chai, Yingyue, Yi Sui, Xinyuan Zhang, Shang Xie, Yifan Kang, Yanrui Feng, Xiaofeng Shan, and Zhigang Cai. 2026. "A Comprehensive Adenoid Cystic Carcinoma-Derived Organoid Platform for Disease Modeling and Drug Screening Captures Interpatient Heterogeneity" Cells 15, no. 4: 383. https://doi.org/10.3390/cells15040383

APA Style

Chai, Y., Sui, Y., Zhang, X., Xie, S., Kang, Y., Feng, Y., Shan, X., & Cai, Z. (2026). A Comprehensive Adenoid Cystic Carcinoma-Derived Organoid Platform for Disease Modeling and Drug Screening Captures Interpatient Heterogeneity. Cells, 15(4), 383. https://doi.org/10.3390/cells15040383

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop