5.2.1 Model Establishment

Orthotopic cancer models are now becoming popular for testing antitumor drug efficacy (Hackl et al. 2013; Day et al. 2015; Hoffman 1999). Compared with the subcutaneous tumors, the biggest advantage of the orthotopic model is that the tumors are grown at the “correct” anatomical site. Implantation of tumor cells into the organ of origin (orthotopic) allows disease‐specific interactions between tumor cells and surrounding stroma (Devaud et al. 2014). It has been shown that orthotopic tumors have different growth, differentiation, metastasis, and drug sensitivity compared with the subcutaneous tumors established by the same tumor cell line (Zhao et al. 2017; Hoffman 2015, 1999; Manzotti et al. 1993). In the past few decades, several orthotopic CRC models have been established and reported (Zhao et al. 2017; Fu et al. 1991; Rashidi et al. 2000; Hiroshima et al. 2014; Zhao et al. 2021; Westcott et al. 2021). However, due to the technical difficulty and a lack of consistency in the model establishment, only the surgical tumor implantation in the cecum and endoscopy‐guided intra‐colon wall tumor cell injection are well studied and used by different investigators.

5.2.2 Surgical Tumor Implantation in the Cecum

Establishing a surgical orthotopic CRC model is technically challenging because the colon in a mouse is a relatively deep organ with an extremely thin muscularis layer. Unlike the human cecum, which is only a very small portion of the large intestine, the cecum in a mouse represents up to one‐third of the large intestine (Figure 5.1a,b). The mouse cecum is a large, thinly walled, blind, and pouch‐shaped organ that connects the ileum and colon. Compared with the distant mouse colon, the dilated, pouch‐shaped cecum has a larger surface area and lower tension. The separable serosa layer of the cecum provides a relatively isolated space for tumor implantation. The dense network of capillaries on the surface of the cecum provides plenty of blood supply to support tumor growth. These features make the cecum an ideal space for implanting tumor tissues or inoculating tumor cells. For those reasons, this model is technically challenging for researchers who are not well versed in small animal surgery. Before the development of endoscopy‐guided intra‐colon wall tumor cell injection, the cecum model was almost the only widely used orthotopic CRC mouse model (Rajput et al. 2008; Céspedes et al. 2007; Lupu et al. 2006). In the labs that are not equipped with small animal endoscopy, the cecum model is still the first choice.

5.2.3 Endoscopy‐Guided Intra‐colon Wall Tumor Cell Injection

Using a small animal endoscopy provides a more technically feasible and less invasive method to perform the orthotopic tumor cell injection as well as to perform post‐surgery monitoring and to obtain biopsies of the mouse colon (Figure 5.1c). In recent years, orthotopic injection of tumor cells into the colorectal wall by endoscopy has been standardized (Zhao et al. 2017, 2021; Westcott et al. 2021). The detailed procedures were reported in our previous publications (Zhao et al. 2017, 2021). By taking advantage of high‐resolution digital imaging, endoscopy provides an enlarged colorectal lumen image. A live feed through the endoscopic camera helps locate the best tumor cell injection site and guides the injection process to ensure successful injection. Based on our experience, with proper training and practice, an experienced researcher can achieve over 70% success rate in the orthotopic tumor cell injection.

Comparison of the endoscopy‐guided intra‐colon wall tumor cell injection and surgical cecum tumor implantation reveals the specific features of each model. The cecum model has the largest area for implantation of tumor tissue chunks, which have the original tumor microenvironment. Meanwhile, because the cecum is a relatively isolated part of the intestine, partial resection of the cecum with a tumor is easy and does not require end‐to‐end anastomosis. Compared with the cecum model, the endoscopy‐guided intra‐colon wall tumor cell injection is a minimally invasive procedure that will not stimulate the regional and systemic immune response, which avoids the potential compounding effect of surgeries on antitumor immunity. However, successful orthotopic injection must place the tumor in between the mucosa and muscularis layers. Shallow implantation (closer to the mucosa) can cause the tumor cells to leak out into the lumen, while deep implantation (closer to the muscularis layer) can potentially perforate the colon and cause the animal to die. This model thus requires extensive training and practice.

5.2.4 Immune Features

Different organs have diverse regional immunity and therefore will influence the antitumor immune responses. Our previous study has compared the antitumor immune response in orthotopic CRC models and subcutaneous CRC models (Zhao et al. 2017, 2021). To avoid the surgical effects on antitumor immunity, we performed the endoscopy‐guided intra‐colon wall tumor cells injection to establish an orthotopic tumor in the colon. The same mouse cell line was used to establish a subcutaneous model. We noticed that orthotopic tumors are more immunogenic than subcutaneous tumors. Specifically, more T‐cells, natural killer cells, and fewer myeloid‐derived suppressor cells are observed in orthotopic tumors than in subcutaneous tumors. The concentration of key cytokines, such as granzyme B, IL‐2, and interferon‐γ is higher in orthotopic tumors as well. Due to the different immunological features, the two models showed distinct sensitivity to immune checkpoints inhibitors. This study indicated that the orthotopic tumor models have different biological characters from subcutaneous models. For those reasons, we think the subcutaneous CRC model is more likely to mimic human diseases with poor immune cell infiltration, whereas the orthotopic model is more similar to human tumors with stronger basal immunity.

5.2.5 Metastasis Features

The subcutaneous connective tissue is a relatively isolated compartment from the deep organs. CRC cell lines, both human‐ and mouse‐derived, are unlikely to form metastasis when injected into the subcutaneous tissue. However, when the same cell line is used to establish the cecum orthotopic model, extensive metastases are observed (Fu et al. 1991; Rashidi et al. 2000; Rajput et al. 2008; Céspedes et al. 2007). With tumor progression, the tumor‐draining lymph nodes become positive of tumor cells. The tumors that break through the barrier of the cecum wall can seed to the abdominal cavity and form extensive peritoneal metastasis. Metastases in the liver and spleen are also reported (Fu et al. 1991; Rashidi et al. 2000; Rajput et al. 2008; Céspedes et al. 2007).

Human CRC treatment is largely dependent on the stage of the disease. Early‐stage diseases without metastases have a great chance to be cured by surgical resection. However, for advanced diseases with extensive metastases, the current treatments are very limited. The metastatic potential of cecum‐based CRC models, therefore, provides a powerful tool to study preclinical treatments for advanced CRC.

5.2.6 Advantages and Limitations

The orthotopic CRC models provide a valuable platform to study cancer progression and tumor response to novel treatments. Placing the tumor in its native environment eliminates many variables, while also adding the proper microenvironment, including the microbiota for CRC. Situated in their native environment, the orthotopic tumors exhibit many similarities with human tumors, thus making this model more realistic to study human disease. However, as discussed earlier, this model is difficult to establish, requiring either specialized equipment (small animal endoscopy) or technical expertise in rodent surgery.

The biological characteristics of orthotopic tumors are largely determined by the tumor implanted. Recent developments in tumor organoids provide another option for establishing orthotopic CRC models (Westcott et al. 2021; O’Rourke et al. 2017; Roper et al. 2017; Drost and Clevers 2018; Crespo et al. 2017). The implanted tumor organoids that are derived from spontaneous mouse CRC recruit a tumor microenvironment from the surrounding cecum or colon tissue and form tumors that mimic human cancer genomic, anatomic, and histological features (Westcott et al. 2021; O’Rourke et al. 2017; Roper et al. 2017; Drost and Clevers 2018; Crespo et al. 2017). The orthotopic implantation of tumor organoids is better for preclinical study than the germline genetic model since the implantation ensures that all mice in the same experiment have the same number and similar volume of tumors. Those advantages will make the organoid‐based orthotopic CRC model the most attractive and valuable model for future preclinical studies.