Bioluminescent Tracking of Human and Mouse Acute Lymphoblastic Leukemia Reveals Potent Immunogenicity of Luciferase In Some Preclinical Models of Leukemia

Abstract 2140

Xenograft models have quickly become the preferred methodology for the preclinical evaluation of treatments for acute lymphoblastic leukemia (ALL). The efficient engraftments in immune-deficient mice achieved with both primary ALL samples and cell lines have facilitated identification of the anti-ALL activity of a wide variety of agents. Despite widespread usage, however, little is known about the early ALL localization and engraftment kinetics in this model, limiting experimental read-outs primarily to survival and end-point analysis at high disease burden. In this study, we have developed bioluminescent imaging of ALL cells to provide a noninvasive, longitudinal measure of leukemia burden that will enhance the sensitivity of preclinical models. Three human precursor B cell (BCP) ALL lines (Nalm-6, RS-4-11 and 380) and two murine BCP ALL lines (289 and 309) were stably tranduced with a lentiviral vector conferring expression of both green fluorescent protein (GFP) and firefly luciferase (ffLuc). Non-obese diabetic/severe combined immunodeficient/IL2Rgamma null (NSG) mice were injected intravenously with 1×10⁶ ALL cells via the lateral tail vein and imaged daily for the first 7 days, then twice weekly thereafter. Animals were also monitored weekly for peripheral leukemia burden by flow cytometric detection of GFP positive cells in blood. Each human ALL line was readily detectable by bioluminescence within 48 hours of injection, providing a measure of disease burden at least one week earlier than can be achieved by peripheral disease monitoring. The human ALL lines Nalm-6 and RS-4-11 initially concentrated in the liver and bone marrow of NSG mice, only appearing in the spleen after 1–2 weeks, while 380 first localized to bone marrow only. In contrast, the murine ALL lines were rapidly detectable in spleen and bone marrow but did not accumulate in the liver. For both murine and human ALL, the initial localization was followed by in situ expansion and subsequent seeding of peripheral sites, with disease burden correlating to increasing bioluminescence intensity. This study, therefore, reveals significant cell line- and species-related differences in leukemia migration, especially early in expansion, which may confound observations between various leukemia models. Furthermore, in a pilot study we demonstrate that this in vivo imaging approach is feasible for primary human ALL samples. To evaluate the utility of bioluminescence in an immune competent leukemia model, we compared the engraftment of ffLuc/GFP+ mouse ALL in syngeneic wild-type (wt) and immune-deficient mice. In contrast to the unhindered rapid expansion of disease in NSG and syngeneic (H-2d) gc-/- (lymphocyte deficient) mice (median survival 21 days, p<0.05 versus wt), wild-type mice sustained a low level of disease for the first 7 days that was subsequently eliminated. Unlabeled and GFP-only+ ALL cells engraft and expand rapidly in wt mice (median survival 25 and 18 days, respectively), and NK-replete/T and B cell-deficient mice engraft with ffLuc/GFP+ ALL cells after an initial delay in expansion (median survival 25 days), indicating that ffLuc is the target of an immune response. This is further supported by a competitive repopulation experiment in which wt mice received 1×10⁶ mixed population cells (95% ffLuc/GFP+ cells and 5% unlabeled leukemia); no mice developed ffLuc/GFP+ disease, while 4/9 eventually developed unlabeled disease. Overall this study demonstrates the increased sensitivity and potential for standardization that in vivo bioluminescent imaging confers on xenograft ALL models. The application of this bioluminescence approach, however, will be limited in immune competent ALL models by the strong immune-mediated clearance of ffLuc+ cells.