Organoid of ovarian cancer: genomic analysis and drug screening

Cell lines are special types of organizations with unlimited proliferation potential on plastic Petri dishes. More than 50 OC cell lines have been established and used in preclinical research [6], and the common ones are SKOV-3, A2780, OVCAR-3, CAOV3, and IGROV-1 [7]. Cell lines are simple, easy to operate, usually used to screen antineoplastic drugs and histological subtypes of OC in vitro to study anticancer pathways, and easily employed to conduct large-scale and high-throughput drug screening. For example, cell lines can be used in investigating the effects of genes on cell survival, proliferation, and kinase-specific inhibitors because of their immortality and accessibility [8].

However, these cell lines have been passed on for several generations or even decades in vitro. Owing to highly evolutionary selection, the genomic structure of cell lines inevitably changes. Stromal cells and extracellular matrix proteins of tumor tissues are lost during in vitro passage, and only the cells that have the same phenotype and successfully adapted to the environment can survive, resulting in the loss of tumor heterogeneity [9]. As a result, these cell lines have different gene expression profiles compared with primary tumors and thus cannot accurately represent the genotype and pathological characteristics of the latter. The advantages of OC cell lines have rendered their involvement in OC research. However, the reliability of cell line source data has been questioned due to unsatisfactory results from clinical trials. Primary cells are still the choice of many researchers.

The commonly used in vivo animal models include the cell line allotransplantation model and PDX. The former transplants cell lines into immunodeficient mice and has been widely used in OC drug screening. Cell transplantation methods include subcutaneous, in situ, and intraperitoneal inoculations. Considering that precision medicine emphasizes the importance of patient-derived models, researchers have upgraded cell line xenotransplantation models and called them PDX. Fresh tumor cells or tissues of resected patients were implanted into the immunodeficient mice, PDX, to preserve genome integrity and tumor heterogeneity [10, 11]. PDX transplants the patient’s tumor fragments into immunodeficient mice and can be used for resected tissue, biopsy, ascites/pleural effusion, and pleural sampling. Subcutaneous inoculation is the main method of inoculation. PDX has been established from various types of cancers [12] and has advantages, such as retention of the interaction between tumors and tissues and good representation of the nature of tumors. In the established PDXs of melanoma, breast cancer, and pancreatic cancer, the chemotherapy response is similar to that from the patient’s source and could accurately reflect the natural state of the tumor [13, 14]. Cytogenetic analysis revealed that compared with cell lines, PDX retains the chromosome structure of tumor tissues [15]. However, PDX use is limited in the study of ovarian cancer [16, 17]. At present, this model is only applied in the early stage of model establishment and chemotherapy response observation. Colombo’s team [18] established 35 large-scale ovarian cancer PDXs. The pathological types include clear cell carcinoma, mucinous carcinoma, and carcinosarcoma. Ricci’s team [19] established some special PDXs of ovarian cancer, such as recurrent tumor and Akt and ERK pathway platinum-resistant tumor, which verified the efficacy of antitumor multi-drug (paclitaxel, bevacizumab, and MEK inhibitor). The results show that the combination of three drugs provides better tolerance and antitumor effect than the combination of two drugs. In addition, PDX is an important tool for the development of new second-line antitumor drugs. Another team established a series of ovarian cancer PDX that banks with different cisplatin sensitivity. The results confirmed that the PDX of ovarian cancer provides a platform for DNA damage repair test and is an important tool in evaluating the response of patients to cisplatin treatment [20].

Ovarian cancer PDX shows a good application prospect and is beneficial for the analysis of ovarian cancer heterogeneity, drug screening, individualized treatment, and recurrence analysis. Furthermore, PDX helps improve clinical efficacy, high recurrence, and early diagnosis rate. However, compared with the simple monolayer cultures of cell lines, PDX is time consuming, expensive, and has a relatively low success rate. Moreover, the effect of immunity on tumors is ignored, and the evaluation of pharmacodynamics and toxicity is limited because only immunodeficient mice are used. Thus, improving PDX to study the immune mechanism of tumors is necessary. Another disadvantage is that cell lines and PDX share the same characteristics. Both easily cultivate tumors with high malignancy, but have difficulty in cultivating tumors with low malignancy [21‐23]. These disadvantages have hindered the wide application of PDX.

Relative to the 2D structure of cell lines, the globule is a globular cell aggregate that remains as a floating 3D structure. Compared with cell lines and PDX, spheroids can greatly restore the microenvironment of tumors and maintain similarity to the original tumors [24‐26]. Ovarian cancer development includes the separation of cancer cells from the fallopian tube or carcinoma in situ, their proliferation to the peritoneal cavity in the form of single cells or globules, and their attachment to the epidermis of the greater omentum, large intestine, and abdomen [27, 28]. Researchers began to explore the spheroids derived from cancer ascites. Malignant ascites, which contain tumor cells suspended into a sphere and are rich in ovarian cancer stem cells [29, 30], are a challenge for ovarian cancer treatment. Functional enrichment of stem cell-like cell populations can be achieved through globular cultures. The spheroid culture is used in studying drug resistance and metastasis of cancer due to the cancer stem cell theory, which holds that the malignant phenotype of cancer is mainly mediated by the stem cell-like part [31‐33]. Detailed analysis of ovarian cancer spheroids revealed the mutual regulatory pathway of ALDH1 and Sox2 involved in ovarian cancer stem cells [34]. Another study group tested the sensitivity of cisplatin, ALDH inhibitor, and JAK2 inhibitor by detecting the ALDH1 and CD133 of globular stem cells. The results showed that globular stem cells sustain their resistance to cisplatin/ALDH inhibitor, although the expression of ALDH is low, and that of CD133 is completely lost. However, spheres resistant to cisplatin/JAK2 inhibitors have enriched ALDH+91 cells [35]. In the above-mentioned studies, the culture of spheroids originated from the ascites in the late stage of cancer. However, the globule cannot cultivate the tissues of normal fallopian tube (FT) and ovarian epithelium (OSE), resulting in a lack of control group for comparison. Moreover, the destruction of cell–cell adhesion can easily induce the dysfunctional apoptosis of epithelial cells and consequently reduce the success rate of OC globular model culture [7].

The organoid is a 3D tissue model directly induced by stem cells and a newly emerged preclinical model. Its application has been extended to many fields. At present, the organoid can be produced from primary prostate, colon, and pancreatic cancers [36‐38] (Table 2). The OC organoid can multiply normal and precancerous cells, and its success rate is higher than those of PDX and spheroid [7].

Organoids can be cultured from different tissues of different patients, such as primary and metastatic tumors, blood tissue, ascites, pleural effusion drainage, and normal FT and OSE [90]. Hans Clevers’ team collected 56 organoids from 32 different patients, covering almost all epithelial OC subtypes [90]. Patients with OC are often diagnosed in their advanced stage and sometimes have metastatic lesions. Therefore, multiple organoids can be established in different tissues of patients, proving that this model has good response to the heterogeneity of primary tumors. The team used two OC media for organoid derivation: one with Wnt-conditioned medium (OCwnt medium), and the other without “OCwnt medium.” After two to three generations, a suitable medium for OC organoid culture will appear and thus can be selected. This method improves the success rate of organic matter culture. Moreover, the absolute number of cells recovered from the 3D culture can be increased remarkably by modifying the Matrigel bilayer organoid culture (MBOC) [91, 92] to cope with the digestive resistance and aggregation tendency of OC, and the success rate of organ-like culture can be increased remarkably from 45 to 90% [93] (Fig. 1). The cultured OC-like organs can be cryopreserved, cultured into spheres, or mixed with matrix gel to subcutaneous or orthotopic transplantation to produce PDX for drug sensitivity study in vivo.

The occurrence and development of OC are related to many genes. Studying the changes in OC gene expression profiles through gene analysis is helpful in uncovering disease etiology and finding treatment options. Gene analysis points the way for OC early diagnosis, drug resistance research, and gene therapy. As a representative of primitive tumors, the organoid could be an important tool for gene analysis.