Background: Colorectal cancer is a leading cause of cancer-related death. Small animal models allow for the study of different metastatic patterns, but an optimal model for metastatic colorectal cancer has not been established.
Objective: The purpose of this study was to determine which orthotopic model most accurately emulates the patterns of primary tumor growth and spontaneous liver and lung metastases seen in patients with colorectal cancer.
Design: Using luciferase-tagged HT-29 cells coinoculated with lymph node stromal analog HK cells, 3 tumor cell delivery models were compared: intrarectal injection, intracecal injection, and acid enema followed by cancer cell instillation. Tumor growth was monitored weekly by bioluminescent imaging, and mice were sacrificed based on primary tumor size or signs of systemic decline. Liver and lungs were evaluated for metastases via bioluminescent imaging and histology.
Settings: The study was conducted at a single university center.
Main outcome measures: Primary tumor and metastasis bioluminescent imaging were measured.
Results: Intrarectal injection had the lowest mortality at 4.0% (1/25) compared with the intracecal group at 17.4% (4/23) and the acid enema followed by cancer cell instillation group at 15.0% (3/20).The primary tumors in intrarectal mice had the highest average bioluminescence (3.78 × 10 ± 4.94 × 10 photons) compared with the mice in the intracecal (9.52 × 10 ± 1.92 × 10 photons; p = 0.012) and acid enema followed by cancer cell instillation groups (6.23 × 10 ± 1.23 × 10 photons; p = 0.0016). A total of 100% of intrarectal and intracecal mice but only 35% of mice in the acid enema followed by cancer cell instillation group had positive bioluminescent imaging before necropsy. Sixty percent of intrarectal mice had liver metastases, and 56% had lung metastases. In the intracecal group, 39% of mice had liver metastases, and 35% had lung metastases. Only 2 acid enema followed by cancer cell instillation mice developed metastases.
Limitations: Tumor injections were performed by multiple investigators. Distant metastases were confirmed, but local lymph node status was not evaluated.
Conclusions: Intrarectal injection is the safest, most reproducible, and successful orthotopic mouse model for human colorectal cancer primary tumor growth and spontaneous metastasis.