Progress in the application of organoids to breast cancer research

Abstract Breast cancer is the most common cancer diagnosed in women. Breast cancer research is currently based mainly on animal models and traditional cell culture. However, the inherent species gap between humans and animals, as well as differences in organization between organs and cells, limits research advances. The breast cancer organoid can reproduce many of the key features of human breast cancer, thereby providing a new platform for investigating the mechanisms underlying the development, progression, metastasis and drug resistance of breast cancer. The application of organoid technology can also promote drug discovery and the design of individualized treatment strategies. Here, we discuss the latest advances in the use of organoid technology for breast cancer research.


| INTRODUC TI ON
Breast cancer (BC) is the most common invasive cancer in women. 1 It affects approximately 12% of women worldwide, 2 and in 2012, it accounted for 25.2% of cancers diagnosed in women, making it the most common female cancer. 3 BC is usually treated with surgery, which may be followed by chemotherapy, radiation therapy or both; a multidisciplinary approach is preferable. 4 Current research on BC focuses on understanding its genesis and development, and on elucidating metastasis and drug resistance mechanisms. The design of individualized treatment strategies for patients with BC is an important topic. Research on tumour biological behaviour has remained at the level of traditional tumour cell lines and animal models despite the gap between different species and between in vitro and in vivo environments. This has become a stumbling block in the application of increasingly sophisticated high-throughput genomics to clinical research.
In 1907, Henry Van and Peters Wilson demonstrated that mechanically isolated sponge cells can regroup and self-organize to produce a whole organism. 5 The subsequent development of cell biology revealed the existence of stem cells, which can differentiate into various types of cells. 6 The emergence of stem cell biology demonstrated the potential of stem cells for organogenesis in vivo.
Stem cells can form teratomas or embryoids, and differentiated cells organize into different structures comparable to those found in multiple tissue types. 6 The differentiation and transformation of stem cells from a two-dimensional (2D) to a three-dimensional (3D) culture system, which enables the development of complex 3D organ structures, led to the emergence of the organoid field. 6 Since 1987, different 3D culture systems have been developed, and different types of stem cells are used for producing organoids mimicking many organs. An organoid is an in vitro model derived from stem cells.
After 3D system, organoids contain a variety of cell types and can self-organize in a manner similar to their in vivo behaviour by proliferation and differentiation. As a result, they form structures that retain the original organ identity in vivo. 7 The organoid technique first employed organs with abundant epithelial structures, such as the stomach, 8 small intestine, 9 colorectum, 10 pancreas, 11 breast 12,13 and prostate. 14 Recently, organoids have become a new trend for studying the evolution of tumours and evaluating the efficacy and toxicity of drugs. This is because organoids have unique characteristics that allow them to reveal most of the tumour properties at the in vitro level. Organoids can be used for exploring the role of cancer stem cells and tumour metastasis mechanisms, as well as for studying the biological characteristics of tumour cells accurately. 15 This review focuses on the application of organoids to BC research.

| THE ORI G IN OF THE MAMMARY ORG ANOID
From the 3D culture models of normal mammary epithelial cells to the establishment of 3D culture system supporting the growth of human breast primary epithelial cells, the culture of mammary organoid has also experienced gradual development. 13,16,17 The latter facilitates the growth of morphologically complex and hormonesensitive mammary tissues. The primary human epithelial cells were self-organized and showed complex vessels and lobular morphologies in the tissue. The ability to culture hormone-sensitive human mammary tissue in hydrogels with defined components will promote the development of human mammary gland (HMG)-based research, which has potential implications for understanding the biology of mammary cancer. 17 In 2017, Qu et al described a method for generating human mammary-like cells from induced pluripotent stem cells (iPSCs). Human iPSC (hiPSC)-derived mammary-like organoids can be used for establishing in vitro models to elucidate the precise effects of various factors on breast cell transformation and BC development, as well as for personalized bioengineering of breast tissue. 18

| Developments in breast stem cell research
The leucine-rich repeat-containing G protein-coupled receptor 5 (Lgr5) is widely known as a stem cell marker in multiple mammalian tissues, such as the stomach, intestine and skin. [19][20][21][22] Lgr5, the receptor for R-spondin proteins, is a Wnt-mediated signal transduction agonist through the β-catenin/TCF pathway. 21 25 Wang et al reported that Lgr5-rather than Lgr5+ cells form colonies in 3D culture. 26 Zhang et al isolated Lgr5+ cells from the mammary glands of Lgr5-lacZ mice and established breast organoids. The colonies from a single Lgr5+ cell spontaneously form a ductal structure surrounded by basal cells. The lumen cells are arranged in a manner resembling the normal ductal structure of the breast. Lgr5+ cell-derived organisms are sustainable during long-term passage; however, although Lgr5-cells expand into primary colonies, the efficacy of colony formation decreases immediately after passage. In addition, reproductive hormones induce epithelial cell proliferation, leading to a significant increase in lumen diameter, accompanied by squamous cell differentiation. Taken together, these findings support the use of mammary Lgr5+ cells as legitimate mammary stem cells. 27

| Culture of mammary gland organoids in vitro
Induced pluripotent stem cells can be produced directly from terminally differentiated cells. 28 This bypasses the need for embryos, and iPSCs from different individuals can be used to model personalized or patient-specific diseases. HiPSCs can produce a variety of cell types, such as neurons, cardiomyocytes and hepatocytes. 29 Qu et al described a method to generate human mammary-like cells even a mammary gland from iPSCs using a suspension sphere culture system. This method is based on the use of non-neural ectoderm progenitors and a mixed gel floating 3D culture system, which is used to simulate the extracellular matrix (ECM) of mammary gland differentiation. The hiPSC-derived mammary-like organoids can be used to construct in vitro models to examine the effects of various factors on breast cell transformation, breast cancer development and breast tissue individualized bioengineering. 30

| Organoids are an important tool for breast cancer research
The most commonly used models for studying BC are cell lines and patient-derived xenografts (PDX). 31,32 Both model systems have considerable drawbacks, although they have contributed greatly to translational BC research. 33 Tumour cell lines acquire mutations during the culture process, which cannot faithfully simulate the original characteristics of the tumour. In addition, cell culture cannot simulate the interaction between tumour cells and other stromal cells in vivo, as cultured cells are single and lack the hierarchy of different cell types. 34 PDX have a wide range of applications; however, they cannot fully reflect the genetic characteristics and heterogeneity of human tumours. [35][36][37] PDX cannot be used to study the process of tumorigenesis, and tumour xenotransplantation has many limitations, such as time-consuming, laborious, long culture cycle, inefficient and difficult for high-throughput drug screening work. 38 Organoid culture can maintain the original genotype and biological characteristics of the tumour. It has other advantages such as stable passage, relatively simple operation and short culture cycle. Organoid culture technology is very helpful for studying the differentiation of cancer stem cells into different types of tumour cells, revealing the causes of tumour heterogeneity, evolution and metastasis, and for evaluating the efficacy of drugs.
Patient-derived tumour organoids (PDTOs) are pre-clinical models for tumour propagation in vitro. PDTOs provide an excellent platform for the study of tumour progression, invasion and drug responses, as they reflect the cellular heterogeneity present in the primary neoplasm. 39,40 However, organoids are limited by the lack of innervation, blood vessels and immune cells. 41 We compare the differences between the three models in Table 1. Currently, efforts are being made to overcome these limitations. One example is the application of co-culture techniques for organoids and mammospheres.
Co-cultures of Vδ2 + T lymphocytes and organoids derived from primary human mammary epithelial cells have been successful, and these T lymphocytes can effectively eradicate triple-negative breast cancer (TNBC) cells. 42 These findings suggest that T lymphocytes from healthy blood donors can be amplified and activated by organoids and subsequently used to treat patients, as well as offering the possibility of in vitro cytotoxicity tests of T lymphocytes from healthy blood donors to tumours from patients. The results of this study support the organoid as an effective model for the study of tumour progression, invasion and drug responses.

| Gene editing of tumour organoids and establishment of tumour organoid animal models
Early studies showed that breast adenocarcinomas in BRCA-related hereditary breast cancer K14cre; Brca1F/F; p53F/F (KB1P), K14cre; Brca1F/F; p53F/F; Mdr1a/b-/-(KB1PM), and K14cre; Brca2F/F; p53F/F (KB2P) mouse models summarize the key features of human diseases including morphologic, expression of basal labels, genomic instability and hypersensitivity to targeted DNA therapy. [43][44][45] Duarte et al used CRISPR-Cas9 gene editing to cultivate PARPi-naive BRCA1-deficient mammary tumour organoids with the Trp53bp1 mutation and detected the response to olaparib therapy in vivo. Compared with control tumour organoids, which were highly sensitive, transplanted tumour organoids with TRP53BP1 targeting tissue-KB1PM7N.1 showed a limited response to olaparib. The results indicated that the deletion of 53BP1 produced a substantial selective advantage in KB1PM tumour cells, even without PARPi treatment. Moreover, Trp53bp1 frameshift mutations were further enriched after olaparib treatment. Immunohistochemical analysis confirmed the deletion of 53BP1-positive tumour cells. This is consistent with the known role of 53BP1 deletion in PARPi resistance. 46 These results indicate that the CRISPR/CAS9 system can effectively modify GEMM-derived mammary tumour organoids to target genes of interest.

| Research on the mechanism of breast cancer
Connexin 43 (Cx43) gap junctions are generally down-regulated in human mammary cancer tissues compared with the non-neoplastic mammary gland tissue surrounding primary tumours. 47 In addition, both Cx26 and Cx43 are down-regulated in many breast cancer cell lines, indicating that gap junctions play a role in maintaining cell differentiation and preventing transformation. [48][49][50][51] Conversely, when connexin is overexpressed in cancer cells, tumour growth slows down, and the cells regain the ability to form at least some differentiated structures. 52

| Challenges and future directions
The main challenge of pre-clinical cancer research is still to establish a model that can summarize the patient's situation as close as possible and retain the intra-tumour heterogeneity and the tumour environment. PDTOs can be used for high-throughput drug screening and selection of effective drugs or drug combinations and to verify the efficacy of these selected drugs. 18 This pre-clinical model can reflect the response of anticancer therapy and give tailored treatments for patients. For example, the organoids growing from the cancer focus during biopsy can guide the individualized treatment plan of patients who need neoadjuvant chemotherapy and palliative treatment without any additional inconvenience to the patients, while the organoids growing from the healthy tissues can provide general information about the drug toxicity. 71 In addition, organoids grown from the patient's liver tissue can be tested to determine its hepatotoxicity, or potential therapies for cardiotoxicity can be obtained from heart cells. 72 In Optical metabolic imaging (OMI) provides a non-invasive method for cell metabolism measurement by using the ratio of NADH and FAD. The redox ratio can provide reliable metabolic readout, so that the technology is superior to those based on single-molecule fluorescence. 74 Imaging technology has many advantages. High resolution makes it possible to track single cells, which helps to identify resistant populations within organoids. 75 to batch, which leads to the difference between experiments. 13,83 This highlights the need for a standard protocol that must reliably integrate the best conditions for breast cancer organoids.

ACK N OWLED G EM ENTS
The present study was supported by the fund of 'San-ming' Project of Medicine in Shenzhen (grant no. SZSM201612010) and the Shenzhen Health and Family Planning Commission Scientific Research Project (grant nos. 201601023).

CO N FLI C T S O F I NTE R E S T
All authors declare that there is no conflict of interest.