Organoid technology in female reproductive biomedicine

The discovery and development of a new drug is a complex, time-consuming, and costly process accompanied by a high rate of failure [53]. Despite substantial progress made in pharmaceutical research in recent years, only a single drug among thousands of laboratory tested compounds reach the marketplace [54]. The main reason for this high rate of failure is related to the lack of reliable disease and relevant human models, and inaccurate results from animal models [55‐57]. The developmental and reproductive toxicity of a vast majority of about 75,000–85,000 chemical substances in commerce has not been investigated [1]. Despite public concerns regarding the reproductive toxicity of chemicals, the field of reproductive toxicology (repro-toxicology) is in its infancy; therefore, more well-targeted research in this field is needed to better understand and prevent reproductive health risks. New approaches in medicine (precision medicine) focus on individual variations that lead to different patient responses for the same drug [58].

Although animal models are frequently used for drug screening and toxicity studies [59], they are not suitable. There is a need to confirm drug safety in both male and female humans [60] along with high-throughput screening of a wide range of drugs and compounds for different purposes, including development of novel contraceptive agents and vaccines, drug screening for infectious diseases, cancer drug development, gestational drug development and reproductive toxicity testing of drugs and compounds [61, 62]. Although traditional in vitro models such as the 2D monolayer cell culture provide valuable tools for drug screening and toxicity testing and have the potential to identify drug candidates, many challenges still remain. One of the main challenges is the change in cellular responses in these model systems that contributes to their unnatural microenvironment [63]. Conventional models, including animal models and traditional culture systems, are unable to provide a platform for precision medicine. Limitations associated with these models have encouraged development and validation of new in vitro models that mimic the complex and dynamic biological features of human tissues, re-create the function and structure of these tissues, and recapitulate in vivo physiology. In this regard, organoids are 3D miniaturizations of human tissues that exhibit native tissue architecture [64] that carries out person-specific genomic and epigenetic information. Organoids provide several unique advantages for drug screening and toxicology studies. These structures are frequently derived from primary cells. They consist of multiple cell populations, possess stable genotypes, and their capacity for self-renewal facilitates their propagation and expansion for drug screening and toxicity studies [65‐67]. Since the use of organoids in reproductive medicine for toxicology studies, drug development and personalized medicine is still a naïve field, more studies are required to evaluate their potential as model systems.

An impressive proof-of-concept has been reported for organoid technology for drug testing in cystic fibrosis (CF) [68, 69]. CF is a life-shortening disease caused by loss of function mutations in the CF transmembrane conductance regulator (CFTR) gene, which results in an accumulation of abnormally viscous, sticky mucus in the gastrointestinal and respiratory tracts [70]. Females with CF have reduced fertility due to abnormally thick, dense cervical mucus, altered ion and fluid transport throughout the reproductive tract, and endocrine abnormalities or menstrual irregularities [70]. Over the past decade, high-throughput screening has led to the identification of small molecules that restore the defective protein and ameliorate the trafficking defects (correctors) and/or augment its activity (potentiators). However, CFTR-restoring pharmacotherapy has varying degrees of efficacy in treating CF, which suggests a contribution by more factors beyond the genetic background among individuals. Dekkers et al. have introduced a novel functional CFTR assay that used organoid cultures from rectal biopsies. They showed that forskolin induced swelling in the healthy organoids. Although this effect was strongly reduced in organoids established from CF patients, it was reversed by treatment with CFTR-restoring compounds [69]. Results of follow-up studies showed that in vitro drug responses in biopsy-derived organoids positively correlated with the clinical response to therapy [68, 71]. Hence, this platform could be used for preclinical selection of responders to pre-existent CFTR modulators as well as for the identification of novel CFTR-restoring compounds. Furthermore, in a recent study, endometrial cancer organoids exhibited patient-specific drug responses. A specific endometrial cancer organoid line was most sensitive to everolimus (an inhibitor of mammalian target of rapamycin [mTOR]), which suggested a strong dependence on the PI3K-AKT pathway and was in line with mutations in the pathway’s signaling mediators (PTEN, PIK3CA, AKT1) [24]. These evidences have shown that organoids are amenable to toxicology studies, drug development, and personalized medicine.

There are three types of human reproductive tract infections (RTIs) - STDs such as chlamydia, gonorrhea, syphilis, genital herpes, and human immunodeficiency virus (HIV); endogenous infections, which result from overgrowth of organisms normally present in a healthy women’s genital tract; and iatrogenic infections, which are associated with medical procedures such as induced abortions, poor delivery practices or intrauterine device (IUD) insertions [72, 73]. RTIs, generally seen as a ‘silent’ epidemic, are associated with complications of gynecologic and reproductive health, including endometritis, infertility, ectopic pregnancy, PID, chronic pelvic pain, miscarriage, and neonatal blindness. STIs/RTIs also increase the risk of HIV infection and can cause death [74‐76]. The morbidity associated with RTIs affects economic productivity and the quality of life of many individuals and, ultimately, entire communities. Inaccessibility of the internal organs for analysis of the long-term consequences of acute and chronic consequences of RTIs/STDs infections in vivo have encouraged the use of many experimental models such as in vitro, ex vivo, and animal models to study pathogenesis mechanisms, discover biomarkers, and perform drug candidate screenings [77‐80]. Classically, because of the limited life span of primary cell cultures, the majority of studies have been conducted in immortalized cell lines, including HEC-1A cells [9], transformed endometrial line HEC-1B [81‐83], Ishikawa cell line [83], uterine epithelial cell (UECs) line ECC-1 [84], OE-E6/E7 (OEC line) [85, 86], the transformed epithelial cervical line HeLa [87], and HEp-2 epithelial cell line [88, 89]. These cell lines are relatively easy to maintain and have provided important insights into understanding host-pathogen interactions. However, the lack of complexity of these cell types, failure to recapitulate the architecture of in vivo tissues, and inability to produce some tissue-specific factors are main limitations for their use. Thus far, several 3D cell culture systems of human reproductive tissues have been developed that have the potential for use in infectious disease research. These 3D cell culture systems include the rotating wall vessel (RWV) bioreactor [90‐92], organoids [3, 4, 23, 25], and organ-on-a-chip (OOAC) [21] models. A major challenge in infectious disease modeling is the recreation of a 3D microenvironment of native tissues to more accurately model the initiation and progression of an infection. Long et al. [93] and Nickerson et al. [94] have reported the first studies that investigated viral and bacterial infections in 3D models, respectively. In recent years, 3D models have gained increasing interest in the field of infectious disease because of their potential to enhance understanding of disease pathogenesis and in drug screening.

Since organoids exhibit enhanced in vivo-like features, including apico-basal polarization, appropriate localization of cell adhesion molecules, cytokine production, responses to antimicrobials and microbial products, support of commensals, and/or susceptibility to infection, they are theoretically well-suited for infectious disease studies. Various pathogens that have been studied using organoids include Helicobacter pylori (stomach organoids) [95, 96], Salmonella enterica [97] and Clostridium difficile [98] in intestinal organoids, and Zika virus (ZIKV) in human neurospheres and brain organoids [99]. Kessler et al. used human fallopian tube organoids and genital Chlamydia trachomatis (C. trachomatis) serovars D, K, and E for long-term in vitro infection analysis and investigated its effect on epithelial homeostasis [100]. The epithelial organoids responded to the infection with a fast, dynamic extrusion of intact Ctr inclusions and/or infected cells into the lumen, and with compensatory cellular proliferation that demonstrated a role for epithelial cells in the defense against this pathogen. Furthermore, the acute infection led to activation of multiple paracrine growth factors (TGF-β1, EGF, FGF). LIF signaling is a key player in the maintenance of stemness in the organoids. The gradual decrease in the number of ciliated cells and increase in CD24 + Epcam+ and CD133 + cells suggests a sustained shift in the regulation of epithelial renewal during an infection. Infected organoids have a less differentiated phenotype with higher stemness potential. Moreover, their methylation data support a potential role for chronic C. trachomatis infection as an epigenetic modulator. C. trachomatis increases hypermethylation of DNA, which is an indicator of accelerated molecular aging [100]. This organoid approach could contribute to a better understanding of the long-term effects of C. trachomatis infections in the development of tubal pathologies, including the initiation of high-grade serous ovarian cancer.

Along with other infectious models, organoids also have limitations; the site of infection in most pathogens is the apical portion of the epithelium and delivery of the pathogen to the lumen of the cystic organoids is a challenge. To circumvent this limitation, microinjections have been used to deliver pathogens to the luminal side of the organoids.

Organoid technology provided a research platform to study reproductive cancers, including ovarian and uterine cancers. Ovarian cancer is the deadliest gynecologic cancer worldwide with approximately a 35% five-year survival rate [101]. It is reported that epithelial ovarian cancer (EOC) accounts for 85–90% of ovarian cancers and high-grade serous carcinoma (HGSC) represents nearly 70% of all EOCs, which is associated with a poorer prognosis [101]. The majority of HGSCs appear to arise from the secretory cells of the FTE rather than from the ovarian surface epithelium [102, 103]. A better understanding of the HGSC biology, FTE transformation, and initiation and progression of HGSC using relevant in vitro human models can lead to new-targeted therapies and immunotherapeutic approaches. Recently, an iPSC- and patient-derived human FTE organoid in vitro model with the relevant cell types (ciliated and secretory) of the human fallopian tube and luminal architecture that closely mimic the tissue-specific structure has been established [22, 25]. iPSC-derived organoids can help to understand physiological and pathological processes of fallopian tubes, provide a powerful platform for drug screening, and develop novel therapies. This in vitro model can provide an appropriate model to explore the fallopian tube origin of HGSC, investigate early processes in the initiation and progression of HGSC, and study germline mutations and genetic alterations involved in HGSC [25]. While iPSC-derived FTE organoids exhibit many features of in vivo tissue, their application in high-throughput screening remains difficult due to limited culture scalability. Over the past decade, numerous literatures in the field of ovarian cancer support an oncogenic role of Notch signaling in HGSC. Notch signaling may be involved in ovarian cancer initiation, progression, metastasis, resistance to chemotherapy, cancer stem cell activity, angiogenesis, and epithelial-to-mesenchymal transition (EMT) [104]. Kessler et al. have reported that Notch signaling is a crucial regulator of differentiation and stemness in fallopian tube organoids; inhibition of the Notch pathway by addition of DBZ (Notch γ-secretase inhibitor) changes the differentiation pattern and structure of these organoids. Their results showed that 78 of the 274 ‘stem cell signature genes’ were also significantly downregulated in the DBZ-treated organoids. This fallopian tube organoid along with the above findings enabled researchers to better understand tubal epithelium pathology, including its role in ovarian cancer because most models of tubal carcinogenesis postulate that secretory cell outgrowth is the initial step toward malignant transformation [25, 105, 106]. Potential therapeutic targeting of the molecular aberrations and cellular signaling pathways involved in initiation and progression of ovarian cancer using organoid technology may provide novel treatment options for cancer patients [104]. The ability to efficiently introduce specific genetic alterations to epithelial precursor cells or human iPSCs and generate genetically engineered organoids using the CRISPR/Cas9 system will enable researchers to gain further insight in cancer research.

Uterine malignancies represent the most common diagnosed gynecologic malignancy worldwide and the fourth most common malignancy in women [107]. About 95% of these malignancies are endometrial carcinomas, the origin of which belongs to the endometrial glandular epithelium. The mesenchymal component such as endometrial stromal sarcoma or mixed epithelial and stromal tumors comprise the remainder [108]. The common treatment strategy for endometrial cancers includes surgery (hysterectomy with bilateral salpingo-oophorectomy) followed by chemotherapy and sometimes radiation therapy. If fertility is desired, hormonal therapy may be administered. However, in two meta-analyses that reviewed trials of recurrent or advanced endometrial cancer and trials on progestin in the adjuvant setting, found no evidence of substantial benefits from these drugs [109, 110]. During the past decade, our knowledge of the genetic basis for endometrial cancers has increased exponentially. Data from the Cancer Genome Atlas (TCGA) project showed that 373 patients with endometrial carcinomas had frequent mutations in PTEN, CTNNB1, PIK3CA, ARID1A, PPP2R1A, KRAS, MYC, ERBB2, CTNNB1, CCNE1, FGFR3,S OX17, TP53, PTEN, ARID5B, PIK3R1, FBXW7, and POLE [111]. Today, treatment of endometrial cancers is not based on patients’ genetic characteristics because of differences in mutations between patients. Therefore, personalized medicine that integrates data from whole genome sequencing (WGS) and whole exome sequencing (WES) to identify patients with specific cancer-related mutations with the drug screening patient-derived organoid (PDO) models can provide a platform for developing an effective therapeutic strategy for cancer patients. Successful treatment of endometrial carcinoma will require individualization of therapies based on the molecular and/or genetic make-up of the endometrial carcinoma cells. PDO cultures hold promise as an in vitro model for accurate recapitulation of a wide variety of normal and oncogenic in vivo cellular behaviors. To date, 3D organoids have been generated from biopsies and/or surgical resections of cancers of the colon [112], pancreas [113], lung [114], and prostate [115]. After the establishment of human endometrium organoid culture protocols, Turco et al. [3], Girda et al. [26], Pauli et al. [116], Dasari et al. [117], and Boretto et al. [24] have reported that it is feasible to grow organoids from primary endometrial cancer.

Organoids derived from endometrial cancer patients recapitulate the morphology of the primary tumors and show similar immunohistochemistry (IHC) characteristics with the primary tumors. IHC for MUC1 and SOX17 on both tumor and normal organoids has confirmed their glandular origin [3, 26]. The effects of verteporfin (VP), an FDA approved light-activated drug used in photodynamic therapy (PDT) for macular degeneration, was tested on an organoid model generated from tumor cells of type 1 endometrial carcinoma patient tissue [117]. VP-treated organoids had less expression of YAP and phospho-YAP, and higher expression of cleaved caspase-3. This finding suggested that VP induces apoptosis and more inactive YAP in the cells of organoids [117]. A high throughput drug screen that used a comprehensive library of up-to-date targeted agents was performed on uterine carcinosarcoma and endometrial adenocarcinoma organoids derived from patients with similar driver mutations in PIK3CA and PTEN. The results showed an association between genotype and drug response profiles. For the endometrial adenocarcinoma organoids, combined treatment with buparlisib (PI3K inhibitor) and olaparib (PARP inhibitor) was optimal when compared with other combination strategies. In contrast, for the endometrial adenocarcinoma organoid, the combination of buparlisib and vorinostat (HDAC inhibitor) was among the most effective treatments [116]. Results of a recent study of organoids from endometrial diseases indicated that the organoids were created from different grades and progression stages of endometrial cancer with lower efficiency compared to other endometrial conditions (20% for endometrial cancer organoids versus 100% for eutopic endometrial organoids and 70% for hyperplastic endometrial [HYP-O] organoids). An optimized culture condition by reducing the p38i concentration and adding insulin-like growth factor 1 (IGF1), HGF, and lipids enhanced the efficiency of organoids generated from endometrial cancers. Endometrial cancer organoids have morphological heterogeneity. Those derived from low-grade/stage cancers exhibit glandular-like morphology with a defined lumen. However, endometrial cancer organoids derived from high-grade/stage cancers commonly appear to be dense and lack a visible lumen [24]. Microsatellite instability (MSI) involved in the pathogenesis of about 30% of endometrial cancer cases [118] was also observed in the endometrial cancer organoids [24]. CGH array or low-coverage WGS revealed that the large majority of no somatic copy number alteration (SCNA) in primary tumors were retained in the corresponding cancer organoids, and the majority of the genetic alterations in the primary tumors were retained in the organoids after long-term expansion. Interestingly, a considerable number of new substitutions were retrieved in cancer organoids after long-term expansion [24]. Endometrial cancer organoids recapitulated the disease phenotype in vivo. Subcutaneously injected cancer organoids generated a cell mass that recapitulated histological and molecular features of the primary tumor. Orthotopic engraftment of high-grade cancer organoids into the uterine horn generated a large, invasive, and highly proliferative mass that had the potential for peritoneal metastasis [24]. These endometrial cancer organoids offer researchers a way to probe cellular pathways involved in tumorigenesis and provide a unique alternative model for endocrine profile studies, drug sensitivity, and for correlating data with the genetic landscape of individual tumors prior to treatment in humans. Clinical studies are needed to correlate organoid assay results with patient outcomes.

Organoids have been generated from young women with cervical clear cell carcinoma (cCCC), an extremely rare subtype of cervical cancer. The optimized protocol for organoid generation from gynecological tumors was applied to produce cCCC organoids. cCCC organoids were expanded in the laboratory setting by a modified Matrigel bilayer organoid culture. These organoids were successfully cryopreserved and recovered after thawing. A few mutations were identified in cCCC organoids and CCC component following genomic analysis. Two of these mutations were detected both in cCCC organoids and CCC component. Moreover, development of a xenograft was confirmed following cCCC organoid transplantation into nude mice. Spheroids derived from cCCC organoids showed drug sensitivity and proliferative capacity upon exposure to the anti-cancer drugs commonly used for gynecological cancer (paclitaxel, cisplatin, and gemcitabine) [28].

However, additional studies are needed to clearly emphasize genomic heterogeneity and transcriptome patterns between organoids and their tumors of origin before replacing the exciting models with organoids, implement cancer-derived organoids for high-throughput preclinical screenings, and design targeted and personalized therapies.

Endometriosis is a gynecological disorder that affects 10–15% of women worldwide and 30–50% of women with infertility. The disease is characterized by the growth of ectopic endometrial tissue outside the uterus [119]. Recently, the organoids from endometriosis patients in various disease stages (I, II, III, and IV) from both eutopic and ectopic origins have been developed [24]. These organoids displayed genomic stability during long-term expansion in culture and showed the same hormonal receptor expression as the original tissue. Endometriosis organoids injected into the peritoneal cavity generated implants that expressed endometriosis markers [24]. However, the decreased efficiency of organoid generation from endometriosis samples compared to healthy endometrial biopsies is a challenge. Moreover, the ability of endometriosis organoids to respond to hormones has yet to be addressed.