Author of

• 11503

  +

  P2.03 - Poster Session 2 - Technology and Novel Development (ID 151)

  • Event: WCLC 2013
  • Type: Poster Session
  • Track: Biology
  • Presentations: 2
  • Moderators:
  • Coordinates: 10/29/2013, 09:30 - 16:30, Exhibit Hall, Ground Level

  • 11504

    +

    P2.03-001 - Gene Expression Profile of A549 Cells from Tissue of 4D Model Predicts Poor Prognosis in Lung Cancer Patients (ID 104)

    09:30 - 09:30  |  Author(s): D.L. Gibbons
    • Abstract

    Background
    Tumor microenvironment plays an important role in regulating cell growth and metastasis. Recently, we developed an ex vivo lung cancer model (4D) that forms perfusable tumor nodules on tissue that mimics human lung cancer histopathology and protease secretion patterns better than tumor cells grown on a petri dish (2D). In this study, we aim to determine if the gene expression profile of cells grown in the model (4D) is a better predictor of survival in lung cancer patients compared to the gene expression profile of cells grown on a matrigel (3D).

    Methods
    We compared the gene expression profile (Human OneArray v5 chip) of A549 cells, human lung cancer cell line, grown on petri dish (2D) against the same cells grown in the tissue of our ex vivo model (4D) and performed gene ontology (GO) analysis. We obtained gene expression data of A549 cells grown on petri dish (2D) and matrigel (3D) from GEO Accession No. GSE17347 and compared the gene expression profiles. We analyzed the differential gene signature for 4D and 3D in human lung cancer database for survival.

    Results
    Expression array analysis showed that there were 2954 gene probes differentially expressed between 2D and 4D. The analysis showed up-regulation of genes associated with mesenchymal cells (CDH2 and VIM). GO analysis showed up-regulation of several genes associated with extracellular matrix (MMP1, MMP9, MMP10, COL4A1, COL5A1), polarity (DLX2, GLI2, HOXD10, HOXD11), and cell fate and development (PPM1A, SALL1, SOX4, ZEB2, JAG1, SOX2, TP63). Moreover, expression array analysis of the 2D and 3D showed 1006 genes that were differentially expressed, with only 36 genes (4%) having the same expression pattern as differential expression between 2D and 4D. There was no difference in expression of genes associated with mesenchymal features (CDH2 and VIM) between 2D and 3D. Finally, the differential gene expression signature between 2D and 4D correlated with poor survival in patients with lung cancer (n = 1492, Log rank p = 1.5 x 10[-7]) while the differential gene expression signature between 2D and 3D correlated with good survival in patients with lung cancer (n = 1492, Log rank p = 1.7 x 10[-9]).

    Conclusion
    The genes necessary to form a perfusable nodule in the 4D model are the genes that are important in cell-matrix interaction, polarity and cell fate, which are lacking in a 2D model. The gene expression signature in the 4D model correlates with poor survival in lung cancer patients, which may be due to presence of more cells with mesenchymal features in the 4D model compared to 2D culture. On the other hand, the tumor cells grown on 3D model show the genes important in tumor cell interaction with matrix without difference in genes identify cells with mesenchymal features. Thus, there may be fewer cells with invasive properties, which may explain the good survival in lung cancer patients. The 4D ex vivo lung cancer model may be a better mimic of the natural progression of tumor growth in lung cancer patients compared to 3D model.

  • 11505

    +

    P2.03-002 - Circulating Tumor Cells from 4D Model Mimic Metastatic Pattern in vivo and Gene Expression Profile Predicts Poor Prognosis in Lung Cancer Patients (ID 228)

    09:30 - 09:30  |  Author(s): D.L. Gibbons
    • Abstract

    Background
    Circulating tumor cells are tumor cells found in the vasculature or lymphatics of cancer patients that are thought to be responsible for metastasis. Recently, we developed an ex vivo 4D lung cancer model that forms perfusable tumor nodules. We have found that there are live tumor cells in the circulation in the model (CTCs). In this study, we aim to determine (i) the characteristics of these cells in vivo and (ii) whether the gene expression profile signature of these cells predicts survival in patients with lung cancer.

    Methods
    We used 393P and 344SQ mouse lung cancer cells for the in vivo experiment. The 393P cells that were grown on petri dish (2D) did not form metastatic lesions when injected into the flank of 129sv mice while 344SQ 2D cells did form metastatic lesions. Both cells were grown in the 4D model. We measured the size of the tumors and the number of CTCs for 14 days. We analyzed the mesenchymal characterisitics of 2D vs CTCs. We injected the CTCs in the tail vein of the 129sv mice and determined if they formed metastasis. We used A549 cells for the gene expression experiment. We placed the A549 2D cells in the 4D model and measured the size of the tumors and the number of CTCs. We compared the gene expression profile (Human OneArray v5 chip) of A549 2D against the A549 CTCs. We analyzed the differential gene signature in human lung cancer database for survival.

    Results
    All of the cell lines formed perfusable lung nodules in the 4D model by day 2 and formed circulating tumor cells starting day 5. The 344SQ cells grew faster (p = 0.01) and formed more circulating tumor cells (p = 0.8) compared to the 393P cells initially but by day 14 there was no significant difference in tumor size (p = 0.8) and number of circulating tumor cells (p = 0.7). The 393P CTCs were less mesenchymal (lower CDH2 (p < 0.0001), ZEB1 (p = 0.001) and VIM (p < 0.0001)) compared to 393P 2D while 344SQ CTCs were more mesenchymal (higher CDH2 (p < 0.0001), ZEB1 (p < 0.0001) and VIM (p < 0.0001)) compared to 344SQ 2D. When 393P CTCs were injected into the tail vein of 129sv mice, they did not form metastatic lesions while 344SQ CTCs formed metastatic lesions in the lung. Gene expression analysis of A549 2D compared to A549 CTCs showed that there were 2504 gene probes that were differentially expressed. This signature correlated with poor survival in patients with lung cancer (n = 1492, Log rank p = 2.6 x 10[-5]).

    Conclusion
    The circulating tumor cells that were isolated from the ex vivo 4D lung cancer model mimic the pattern of metastasis in vivo. Moreover, the circulating tumor cell signature correlates with poor survival in patients with lung cancer. The circulating tumor cells that are isolated in the ex vivo 4D lung cancer model may be the cells responsible for metastasis.