Intratibial Injection Causes Direct Pulmonary Seeding of Osteosarcoma Cells and Is Not a Spontaneous Model of Metastasis: A Mouse Osteosarcoma Model.

Background: Although metastasis is the major cause of mortality in patients with osteosarcoma, little is known about how micrometastases progress to gross metastatic disease. Clinically relevant animal models are necessary to facilitate development of new therapies to target indolent pulmonary metastases. Intratibial injection of human and murine osteosarcoma cell lines have been described as orthotopic models that develop spontaneous pulmonary metastasis over time. However, there is variability in reported injection techniques and metastatic efficiency.

Questions/purposes: We aimed to characterize a widely used murine model of metastatic osteosarcoma, determine whether it is appropriate to study spontaneous pulmonary metastasis by establishing a reliable volume for intratibial injection, determine the incidence of primary tumor and metastatic formation, determine the kinetics of pulmonary metastatic seeding and outgrowth, and the contribution of the primary tumor to subsequent development of metastasis.

Methods: The metastatic mouse osteosarcoma cell line K7M2 was injected into the tibia of mice. The maximum volume that could be injected without leakage was determined using Evan's blue dye (n = 8 mice). Primary tumor formation and metastatic efficiency were determined by measuring the incidence of primary tumor and metastatic formation 4 weeks after intratibial injection (n = 30). The kinetics of metastatic development were determined by performing serial euthanasia at 1, 2, 3, and 4 weeks after injection (n = 24; five to six mice per group). Number of metastatic foci/histologic lung section and metastatic burden/lung section (average surface area of metastatic lesions divided by the total surface area of the lung) was calculated in a blinded fashion. To test the contribution of the primary tumor to subsequent metastases, amputations were performed 30 minutes, 4 hours, or 24 hours after injection (n = 21; five to six mice per group). Mice were euthanized after 4 weeks and metastatic burden calculated as described previously, comparing mice that had undergone amputation with control, nonamputated mice. Differences between groups were calculated using Kruskal-Wallis and one-way analysis of variance.

Results: The maximum volume of cell suspension that could be injected without leakage was 10 μL. Intratibial injection of tumor cells led to intramedullary tumor formation in 93% of mice by 4 weeks and resulted in detectable pulmonary metastases in 100% of these mice as early as 1 week post-injection. Metastatic burden increased over time (0.88% ± 0.58, week 1; 6.6% ± 5.3, week 2; 16.1% ± 12.5, week 3; and 40.3% ± 14.83, week 4) with a mean difference from week 1 to week 4 of -39.38 (p < 0.001; 95% confidence interval [CI], -57.39 to -21.37), showing pulmonary metastatic growth over time. In contrast, the mean number of metastatic foci did not increase from week 1 to week 4 (36.4 ± 33.6 versus 49.3 ± 26.3, p = 0.18). Amputation of the injected limb at 30 minutes, 4 hours, and 24 hours after injection did not affect pulmonary metastatic burden at 4 weeks, with amputation as early as 30 minutes post-injection resulting in a metastatic burden equivalent to tumor-bearing controls (48.9% ± 6.1% versus 40.9% ± 15.3%, mean difference 7.96, p = 0.819; 95% CI, -33.9 to 18.0).

Conclusions: There is immediate seeding of the metastatic site after intratibial injection of the K7M2 osteosarcoma cell line, independent of a primary tumor. This is therefore not a model of spontaneous metastasis.

Clinical Relevance: This model should not be used to study the early components of the metastatic cascade, but rather used as an experimental model of metastasis. Improved understanding of this commonly used model will allow for proper interpretation of existing data and inform the design of future studies exploring the biology of metastasis in osteosarcoma.