Advanced flow cytometric analysis of nanoparticle targeting to rare leukemic stem cells in peripheral human blood in a defined model system

摘要

Leukemia stem cells are both stem-like and leukemic-like. This complicates their detection as rare circulating tumor cells in the peripheral blood of leukemia patients. Since leukemic stem cells are also resistant to standard chemotherapeutic regimens, new therapeutic strategies need to be designed to kill the leukemic stem cells without killing normal stem cells. In these initial targeting studies we utilized a bioinformatics approach to design an antibody-fluorescent nanoparticle conjugate for targeting to these leukemic stem cells and to minimize targeting to normal stem-progenitor cells. Multicolor flow cytometric analyses were performed on a BD FACS Aria Ⅲ. Human leukemic stem cell-like cell RS4;11 (with putative immunophenotype CD133+/CD24+/-, CD34+/-, CD38+, CD10-/Flt3+) was spiked into normal hematopoietic stem-progenitor cells obtained from a "buffy coat" prep (with putative immunophenotype CD133-/CD34+/CD38-/CD10-/Flt-3-) to be used as a model human leukemia patient. To analyze the model system, digital data mixtures of the two cell types were first created and assigned classifiers in order to create truth sets. ROC (Receiver Operating Characteristic) and multidimensional cluster analyses were used to evaluate the specificity and sensitivity of the immunophenotyping panel and for automated cell population identification, respectively. Costs of misclassification (false targeting) were also accounted for by this analysis scheme. Ultimately, this analysis scheme will be applied to use of nanoparticle-antibody conjugates at therapeutic doses for targeted killing of leukemia stem cells preferentially to normal stem -progenitor cells.