Bioluminescent imaging (BLI) is an optical approach that enables rapid, non-invasive and real-time analysis of various cellular and molecular events in live animals. For over a decade, BLI has been widely used to monitor tumor take, growth, metastasis and treatment efficacy in preclinical cancer research. The technique is based on the light-emitting enzyme luciferase expressed in the live cells as internal light sources that can go through the tissue and be detected externally as the biological readout. Numerous studies have reported BLI applications for internal organ cancers and metastases, however few, in our opinion, underwent careful and rigorous validation to assess reliability and establish best-practice protocols. Our recent applications of BLI to detect xenograft tumors of human cancers and metastases highlight many practical challenges. In a subcutaneous model with LNCaP-AR-luc prostate cancer cells, we observed a linear regression R2 = 0.062 (n = 32; p = 0.056) for BLI readout (15 minute post i.p. D-luciferin for all mice imaged) vs. tumor weight at necropsy. In addition, subcutaneous xenograft tumors inoculated on two flanks within the same mice could vary as much as 10-fold in BLI per gram. In an intraprostatically-inoculated human PC-3-luc prostate cancer model, we observed a linear regression R2 = 0.080 (n = 16; p = 0.229) for BLI readout vs. tumor weight at necropsy. In a subcutaneous model of MDA-MB-231-luc breast cancer cells, correlation between tumor volume and BLI was less linear than that between tumor weight and tumor volume. Preliminary histology examination suggested that extent of central necrosis could not completely account for the poor correlation between tumor weight (all smaller than 1 g) and BLI. Examination of BLI readout kinetics showed considerable individual variability of peak intensity time, confirming findings of others (e.g., Keyaerts et. al, Eur J Nucl Med Mol Imaging, 2008; 35:999 -1007; Sim et. al, Cancer Res 2011;71:686 - 692). We are currently examining differences in cellular expression of luciferase and tumor vascularity as potential contributors to the discordance between BLI readout and tumor weight. In summary, positive BLI readout can be a reliable predictor of the presence of tumor, but careful validation is required to improve its predictive power of tumor burden for specific BLI applications. Grant supports: R21AT005383; R01AT007395.

Citation Format: Yong Zhang, Suni Tang, Jinhui Zhang, Cheng Jiang, Junxuan Lu. To image or not to image by bioluminescence? A question of validation for live animal imaging. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3933. doi:10.1158/1538-7445.AM2013-3933