Interrogating cell heterogeneity and dormancy in prostate cancer disseminated tumor cells Nazanin S. Ruppender1, Lisha Brown1, Roger Coleman2, Bryce Lakely1, Ilsa Coleman2, Christopher Welty1, Roman Gulati2, Peter S. Nelson2, Paul H. Lange1, Eva Corey1, Robert Vessella1, Colm Morrissey1. 1University of Washington, Seattle, WA; 2Fred Hutchinson Cancer Research Center, Seattle, WA
Dissemination of tumor cells to the bone is an early event in prostate cancer (PCa) and is often insufficient by itself for the development of metastases; while PCa cells may be present in bone marrow, not all patients develop tumors in the bone. Therefore, the characterization of disseminated tumor cells (DTC) in bone marrow is important. We hypothesize that profiling gene expression in DTC could identify molecular cues that determine why some DTC are dormant, which dormant cells have the potential to reactivate, and which DTC will remain dormant. Our objectives were: (1) to develop a method toprofile the transcriptome of individual PCa cells ( single cell PCR? How many genes?), (2) use the method to profiletumor cell heterogeneityandmarkers of dormancy (marker?)in individual DTC from patients, and (3) useLuCaP xenograft lines(Functional heterogeneity?) to examine tumor cell heterogeneity and dormancy in vitro. We have demonstrated previously that we can obtain15,441 positive probes (15 K genes, Agilent 44K single cell arrays?)from one PCa cell using a high stringency cut off on Agilent 44K oligonucleotide arrays. Furthermore, using the same high stringency cut off, thesensitivity and specificity between 1 and 10 cells was 0.640 and 0.946 respectively, demonstrating a very low false positive rate. Using this methodology, we are comparing the gene expression profile of 10 individual DTC from each of 5 patients with no evidence of disease four years after radical prostatectomy and 10 individual DTC from each of 5 patients with advanced metastatic disease. Additionally, our laboratory has developed 24 LuCaP PCa xenografts that do not proliferate in typical in vitro cell culture conditions. However, we have developed a technique using reactive bone marrow stroma that allows for in vitro culture and proliferation of these xenografts. Thus, we are comparing individual LuCaP cells in three cellular ‘states’ in vitro(1) dormant, (2) activated, and (3) inactivatedto interrogate tumor cell heterogeneity and identify markers of tumor cell dormancy. Using this new technique for in vitro culture of xenografts, we show that cell contact, not only to the underlyingstroma but also to other tumor cells, is required for proliferation of LuCaP cells in vitro (need to contact stroma & other tumor cells). Furthermore, we show that proliferating, “active” LuCaP cells can be induced to enter a dormant, “inactive” state through lack of cell-cell contact (Dynamic. In conclusion, our studies will address the heterogeneity of single DTC isolated from the bone marrow of patients with PCa, and using both clinical and experimental specimens,identify potential genesassociated with dormancy in PCa.