An in vivo orthotopic canine model to evaluate distribution of intraprostatic injectate: implications for gene therapy and drug delivery for prostate cancer.

OBJECTIVES: Complete and reliable infiltration of the entire prostate is important to improve the efficacy of intraprostatic gene therapy, as well as drug delivery. We sought to evaluate the optimal injection scheme and feasibility of magnetic resonance imaging (MRI) using a surrogate solution to assess the distribution of injectate in an orthotopic canine model. METHODS: Twelve dogs were anesthetized, and laparotomy was performed. Four dogs in each group were injected with a 1:10 dilution of 1% methylene blue and gadolinium-diethylenetriamine pentaacetic acid using 3, 10, and 20-core injection schemes. The dogs subsequently underwent MRI and then were killed. The prostates were harvested, sectioned, and photographed for methylene blue distribution. The cross-sectional area of the stained pathology and MRI slides was calculated, and correlation studies were performed. Statistical analysis was performed using the paired t test and Pearson's correlation test. RESULTS: The gadolinium distribution reflected the methylene blue distribution, with a Pearson's correlation coefficient of 0.97 (P <0.001). The fractional volume distribution in the 3, 10, and 20-core injection schemes was 10.2%, 28.5%, and 30.1%, respectively. The volume of distribution was significantly greater for the 10-core (P = 0.0024) and 20-core (P = 0.0025) injection schemes than for the 3-core scheme. However, no significant difference was found between the 10 and 20-core schemes (P = 0.52). CONCLUSIONS: MRI using gadolinium-diethylenetriamine pentaacetic acid is an excellent modality to evaluate the effective volume distribution of injectate in an in vivo orthotopic prostate model. The 10-core injection scheme seemed to be as good as the 20-core scheme in achieving adequate distribution of injectate within the prostate.