Novel organoid culture condition: modeling fetal-like plasticity in colorectal cancer

A recent study published in Cell Research established a new condition for patient-derived organoid (PDO) culture. The PDO enables long-term expansion and maintenance with fetal-like features, preserves the oncofetal-like state, and reveals FGF2-AP-1 signaling associated with phenotypic plasticity and increased treatment resistance.

Worldwide, colorectal cancer ranks as the third most common cancer and the second leading cause of cancer-related death. According to data from the World Health Organization, it is estimated that by 2040, new cases will increase by 63% to 3.2 million, and deaths will rise by 73% to 1.6 million.¹ Immunotherapy has made significant progress in patients with microsatellite instability-high (MSI-H) tumors.² A phase 2 study showed that MSI-H locally advanced rectal tumors treated with dostarlimab monotherapy achieved a clinical complete response (100%; 95% confidence interval, 74–100)³; however, only ~15% of colorectal cancers and only 3%–5% of rectal cancers are MSI-H. The vast majority (80%–85%) of colorectal cancers are classified as microsatellite stable (MSS), and these tumors are often referred to as “cold” tumors. The use of immunotherapy in patients with MSS tumors is investigational, and historically, these tumors have not responded well to immunotherapy approaches. Overall, the treatment of MSS colorectal cancer has relied on traditional methods such as chemotherapy, radiation therapy, and surgical resection. Therefore, understanding the mechanism of chemo- or radio-treatment resistance in colorectal cancer remains essential.

Tumor cell plasticity has garnered significant research attention over the past decade. Increasing evidence indicates that cancer cell plasticity plays a role in tumor initiation, progression, metastasis, immune evasion, development of treatment resistance, and therapeutic failure.⁴ In colorectal cancer, chemotherapy induces environmental stress, activates fetal-like transcriptional programs, and pushes cancer cells into highly plastic states. Consequently, cancer cells adapt to various environmental pressures, including those from high-grade and treatment-resistant tumors.⁵ A recent study published in Cell Research by Xiong et al. explored the specific role of these fetal-like programs in driving colorectal cancer cellular plasticity, using chemically induced patient-derived organoids (CiPDOs) as a model.^6,7 A major challenge with the conventional patient-derived organoid (PDO) platform is the difficulty of maintaining long-term growth in vitro while preserving their plastic states. Sato et al. developed the initial method for the long-term expansion of colon epithelial organoids and further evaluated the essential components of PDO cultures, such as R-Spondin, Noggin, and Wnt3a.⁸ While the original organoid culturing system has been a valuable model for studying oncology and cell biology for several decades, the model still has inherent limitations. For example, the culture system relies on unstable and expensive recombinant proteins such as growth factors, which often limit the preservation of plastic cellular states in the dominant clones during PDO expansion.⁹ To overcome some of these limitations, Xiong et al. explored a new CiPDO culturing condition, using small-molecule compounds to replace the recombinant proteins.

A four-component medium (referred to as “4C”) has been developed. In this formulation, epidermal growth factor (EGF) activates its receptor EGFR, triggering the MAPK/ERK pathway; CHIR99021, the GSK3/inhibitor, activates downstream Wnt signaling to replace R-Spondin; LDN-214117, the selective bone morphogenetic protein (BMP) signaling inhibitor, regulates stem cell-related gene expression; fibroblast growth factor 2 (FGF2) binds to its receptors FGFRs, triggering downstream signaling pathways that regulate cell behavior. Whole-exome sequencing and inferred copy number variation analyses confirmed that the CiPDOs retained most of the somatic mutations found in their matched primary tumors under 4C culture conditions. According to Xiong and colleagues’ findings, when FGF2 was absent from the culture medium, PDOs failed to be maintained and undergo long-term expansion, although robust PDO formation is retained. FGF2 is recognized for its diverse impact on multiple cellular processes, including cell growth, differentiation, migration, tissue repair, angiogenesis, and embryonic development, and is considered a potential therapeutic target.¹⁰ Notably, FGF2 improved the efficiency of CiPDO formation, extended the long-term CiPDO expansion potential, and upregulated several fetal marker genes, including ANXA1, ANXA3, MSLN, and LY6D.

The CiPDOs maintained under 4C conditions were further confirmed to retain fetal-like and plastic phenotypes. With FGF2 substantially present, the ex vivo PDOs were found to be enriched for expression of CAV2, CTSE, MSLN, CCND1, and F3 — markers of epithelial-mesenchymal plasticity (EMP)-related programs. They also exhibited EMP-associated spindle-like morphological changes, gained invasive capacity, developed resistance to chemotherapy and radiation therapy, and also showed a higher number of metastatic lesions in the orthotopic cecum xenograft model. Furthermore, advanced-stage (stage III/IV) tumor-derived organoids exhibit higher expression of oncofetal state (OnFS) genes compared to early-stage tumor-derived organoids and share a high degree of similarity with parental tumors through transcriptional comparison with primary tumor datasets. Orthotopic xenografts confirmed that the OnFS expression was preserved. Moreover, across patient samples, the integrated data showed that transcriptional targets of the AP-1 family were enriched during the transition toward the OnFS. At the same time, JUND and FOS were upregulated most consistently as AP-1 components. Therefore, as expected, when JUND and FOS in CiPDOs were knocked down by targeted shRNA or FGF2 was withdrawn from the culture medium, the expression of several essential OnFS markers was significantly reduced, indicating that the FGF2-AP-1 signaling promotes the maintenance of the OnFS in CiPDOs. Interestingly, OnFS expression was consistently higher in tumor-derived organoids compared to normal colonic organoids derived from the same patients and maintained under the same conditions.

This study, although interesting and provocative in many ways, raises several questions that require clarification and need further exploration. EGF and FGF2, as recombinant proteins, are essential components of the culture medium and cannot currently be replaced by less expensive, more stable small molecules or compounds. For long-term culture of PDOs, different growth rates among colonies may cause the faster-growing colonies to outcompete the slower-growing ones, leading to a loss of heterozygosity and increased uniformity in organoids over time. More clinically relevant research is necessary to examine how the OnFS-associated fetal-like states evolve in tumors during chemotherapy and/or during radiation therapy and how they respond to the immune system when fetal-like programs are active. FGF2, as a potential drug target, could lead to the development of new drugs or treatment strategies on the PDO platform; however, it requires significant validation prior to being applicable in the clinic for patients with colorectal cancer. In summary, the authors created a novel PDO culturing system with four key components: CHIR99021, LDN-214117, EGF, and FGF2. Using this platform, organoids can be maintained long-term to better study fetal-like transcriptional states in colorectal cancer organoids, identify the OnFS organoid population, understand the activity of the FGF2-AP-1 pathway related to OnFS expression, and explore the associations and potential mechanisms of resistance to chemoradiation treatment.