The failure of many drugs in the clinic - drugs proven to be active in pre-clinical assays - has raised question marks on the predictive power of these models, often based on cell line data. The divergence of these cell lines from the original tumors, and their rapid growth as simple monocultures on 2D plastic were highlighted as potentially causing these issues. Implantation of the cell lines into mice adds complexity to these models, but the crosstalk with mouse stroma also leads to confounding results (eg, HGF, IL-6).
PREDECT, a European consortium funded by the Innovative Medicines Initiative (IMI), has the goal to build and characterise more predictive Oncology model platforms for three pathologies (lung, breast and prostate). Signalling pathways and heterogeneity of these model platforms are compared to each other as well as to primary human tissue using a central TMA-based platform.
For prostate cancer, a 3D tumor growth assay system was established recently using RFP/FLuc-labelled LNCaP, PC346C, or VCaP human prostate cancer cells grown in a matrix of Matrigel, collagen, or a mix of the two, in the presence or absence of GFP/RLuc labelled human stromal cells (WPMY, or an immortalised cancer-associated fibroblast cell line, CAF). Cell densities, ratios, and matrix concentrations/volumes were optimised for cells cultured for 3 weeks. In an approach to improve tumor-stroma crosstalk in vivo, LNCaP cells pre-grown in 3D cultures with or without human stromal cells were implanted orthotopically into SCID mice. Growth of the tumor and stromal cells was monitored by in vivo bioluminescence, and the effect of the co-implantation on the primary tumor and metastasis was analysed via ex-vivo luciferase assays and IHC.
Whereas LNCaP and PC346C cells readily formed spheroids in 3D culture, VCaPs remained as single cells. Growth of VCaPs was facilitated by pre-forming spheroids in vitro, followed by matrix embedding. Addition of stromal cells stimulated growth of not only the tumor, but also the stromal cells in some conditions, as monitored using RFP and GFP live cell imaging. 3D, as well as stromal cell-mediated treatment resistance was observed. Using the fluorescent dyes Hoechst, EdU, and NucView, the proliferative status of individual cells within spheroids could be analysed in situ. Finally, paraffin-embedding of the 3D cultures allowed for TMA analysis and comparison to other models and patient samples.Results will also be shown on the orthotopic growth of LNCaP tumors and the effect of stromal cell co-implantation.
Introducing complexity to cell culture may help to generate more realistic, and thus also more predictive models - in a first step using even standard cell lines.
Citation Format: Suzana Vidic, Norbert Esser, Ronald de Hoogt, Ingrid Verberne, Ira Kogan-Sakin, Yan Stein, Varda Rotter, Michael Barbier, Yolanda Chong, Sabine De Breucker, Karine Smans, Malin Akerfelt, Matthias Nees, Peter King, Ian Hickson, Wytske van Weerden, Ralph Graeser. Complex in vitro and in vivo prostate cancer models for the PREDECT consortium. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2023. doi:10.1158/1538-7445.AM2014-2023