Introduction Skeletal metastases models typically use parameters such as radiographic analysis, tumor volume, tumor weight, and histological analysis to assess tumor burden. With the exception of radiographic analysis, each of these parameters requires sacrifice of the animal - an undesirable practice both for the numerous animals required as well as the necessary termination of study of the particular animal. We hypothesized that bioluminescence imaging would offer a novel, sensitive, non-invasive means to follow the progression of tumors in vivo. A cooled charge coupled device (CCD) camera captures photons emitted from the enzyme luciferase after exposure to its substrate, luciferin, and integrates them over a period of time.
Materials and Methods The luciferase gene was delivered to human osteolytic prostate cancer (PC-3) cells through in vitro retroviral gene therapy. To determine in vitro correlation, increasing numbers of luciferase expressing PC-3 cells were placed in a 96 well culture plate. At 24 hours, the cells were lysed and a luciferase assay was performed. A correlation coefficient was determined by linear regression. To induce bone lesions in SCID mice, 1 × 105 cells of luciferase expressing PC-3 cells were implanted into the proximal tibias of 8 week old mice. A 3 mm medial parapatellar incision was made, and cells were injected with a 27 1/2 ga needle. Radiographs and CCD images were obtained every 10 days. Photons emitted from PC-3 cells were collected and integrated. Images were obtained by superimposing gray-scale photographic images and luciferase color images. Tumor alone was compared to two different tumor suppression treatments previously shown to be effective for treating osteolytic prostate cancer. Animals were sacrificed at 50 days.
Results A correlation of R2 = 0.95 was found between luciferase activity and cell number in vitro. CCD imaging detected an increase in signal intensity at 10 days, whereas radiographs were unable to detect a lesion until 20 days. Tumor suppression was so significant in one treatment that no lesion was detectable on radiograph, though tumor progression could still be followed with CCD imaging.
Conclusion CCD imaging is a novel, safe and effective means to follow tumor progression in vivo in a skeletal metastasis model. It was more sensitive than radiograph for showing small lesions, and it was able to demonstrate the effectiveness of two previously validated treatments of osteolytic prostate cancer lesions more effectively than conventional radiographs.