| Cancer constitutes an enormous burden on society, with about 14.1 million new cases and 8.2 million cancer deaths according to the GLOBOCAN estimates. And in China, according to the China Health Statistics 2013, cancer is still the leading cause of death, with the mortality rate of 164.51 per 100 thousand. Most cancer death is directly due to invasion and metastasis, which are 2 fundamental properties of cancer biological behaviors, but the underlying mechanisms of cancer invasion remain elusive. Recently, the oncology community has come to the understanding that cancer is a disease of imbalance, in which the fundamental importance of such theoretical changes is that we have to pay particular attention to tumor microenvironment in addition to cancer cells, as tumor microenvironment plays a crucial role via the co-evolution of cancer cell and tumor stroma. However, tumor microenvironment is a complex and dynamic community, which is undergoing constant evolutions during cancer invasion. There are few techniques that can simultaneously image multiple components in a complex tumor microenvironment of the same clinical tissue section. Thus, it is of great urgency to develop new techiniques to image and acquire the essential and hidden information from tumor microenvironment.Quantum dots (QDs), with its unique size and surface effects, presented the features such as size-and composition-tunable light emission, enhanced signal brightness, resistance to photobleaching, which have shown great potential in biomedical application, especially in multiplexed imaging. In addition, taking the advantages of the combination of multispectral imaging analysis and computer-based algorithm, a new platform based on such imaging and analysis system could be established to reveal the information on co-evolution between cancer cells and tumor microenvironment from both morphological and statistical aspects.Considering the current clinical dilemma, this study was aimed to develop a new platform to explore the mechanism of cancer invasion and metastasis from both cancer cells and tumor microenvironment. We firstly established standard QDs-based multiplexed imaging system to simultaneously image the key components in tumor microenvironment. Then, combined with the multispectral imaging analysis system, the morphological and tempo-spatial information between cancer cells and tumor microenvironment could be further demonstrated at the architectural level. Furthermore, with the aid of computer-based algorithm, the quantitative and statistical information could be explored to evaluate the prognosis performance. As expected, the combined features with cancer cells and tumor microenvironment could open a new frontier for both basic cancer research and clinical application.Three studies have been conducted in this research project:Part I:In vitro three-dimensional cell culture study by QDs-based molecular imaging Objective This study was aimed to establish a new in vitro three-dimensional (3D) cell culture and use QDs molecular imaging techinique to examine the invasive behaviors of hepatocellular carcinoma (HCC) cells.Methods Each well of the 24-well cell culture plate was cover-slipped. Matrigel diluted with serum-free DMEM was added and HCCLM9 cells were cultured on the Matrigel. The cell morphological and cell growth characteristics were observed by inverted microscopy and laser confocal microscopy at different culture time. Cell invasive features were monitored by QDs-based real-time molecular imaging techniques.Results The results showed that on this 3D cell culture platform, HCCLM9 cells exhibited typical multi-step invasive behaviors, including reversion of cell senescence, active focal proliferation and dominant clones invasion. During the process, cells under 3D cell culture showed biological behaviors of spatio-temporal characteristics. Cells first merged on the surface of matrix, then gradually infiltrated and migrated into deep part of matrix, presenting polygonal morphology with stretched protrusions, forming tubular, annular and even network structure, which suggested that HCC cells have the morphological basis for vasculogenic mimicry. In addition, small cell clones with their edges well-circumscribed in early stage, progressed into a large irregular clone with ill-defined edge, while the other cells developed invadopodia.Conclusion These findings indicate that a novel 3D cell culture platform has been successfully established, which can mimic the in vivo tumor microenvironment, and when combined with QDs-based molecular imaging, it can help to better investigate the invasive behaviors of HCC cells.Part II:Dynamic changes of collagen IV during cancer invasion and migrationObjective This study was aimed to establish the standard protocol of QDs-based molecular imaging technique, and further to reveal the mechanisms of cancer invasion based on dynamic changes of collagen IV by novel QDs imaging technique.Methods Cancer tissues of HCC, gastric cancer (GC), breast cancer (BC) and cervical cancer (CC) were collected and stained by traditional immunohistochemistry and novel QDs-based molecular imaging technique. Several key molecules representing of cancer cells and tumor microenvironment were studied.Results During cancer invasion and migration, collagen ? underwent structural and functional changes. First, collagen ? increased at basement membrane, presenting an irregular sheath surrounding cancer nests; then, collagen ? was degraded to form invasion fronts at several sites accompanied with linear re-deposition of collagen ? in extracellular matrix (ECM). Finally, cancer cells escaped from large cancer nests from the degraded parts and along the linear rigid invasion highway.Conclusion During cancer invasion and migration, collagen ? underwent dynamic changes, which could provide a proper tumor microenvironment for cancer metastasis.Part III:The co-evolution between cancer cells and tumor microenvironment by QDs-based multiplexed imagingObjective This part of study was aimed to establish the standard protocol of QDs-based multiplexed imaging technique and further to explore the co-evolution between cancer cells and key components in tumor microenvironment, majorly including tumor associated macrophages (TAMs) and tumor neo-vessels so as to understand the basic rules of cancer invasion and metastasis in depth, which could help predict the prognosis and guide the cancer treatment strategy.Methods Cancer tissues and responding clinical information of CC and GC were collected. The standard imaging and analysis system was established by QDs-based multiplexed imaging technique, multispectral analysis system and computer-aided algorithm. And further the co-evolution information between cancer cells and tumor microenvironment was analyzed from morphological and statistical aspects with the help of above analysis system.Results The QDs-based multiplexed imaging and analysis system was applicapable, reproducible, reliable and easy to use for clinical samples. During squamous cancer and ademocarcinoma invasion, the co-evolution of cancer cells and tumor microenvironment presents the same dynamic rules. During the progression process from chronic cervicitis, cervical intraepithelial neoplasia (CIN ?-?) to CC, the median value of TAMs density was higher in the CC group (5540) than the CIN ?-? group (2502) and the chronic cervicitis group (1403), with statistical significance in three groups (P=0.000, for between-group comparisons). The number of neo-vessels was also much higher in the CC group (27) than the CIN ?-? group (17) or the chronic cervicitis group (6.5), with statistical significance in three groups (P=0.000, for between group comparisons). In addition, according to the quantitative analysis of GC, the mean TAMs density was higher in the poorly differentiated group (1544) than the well-differentiated group (317) and the signet-ring cell group (1011) (P=0.001, for between group comparisons), with statistical significance in three groups. The results of tumor neo-vessels analyses were the same with TAMs. The poorly differentiated group has the highest count but the well differentiated group has the lowest count, with statistical significance in three groups (P=0.029, for between group comparisons). And, the number of tumor invasion unit was much higher in the poorly differentiated group (373) than the well-differentiated group (82) or the signet-ring cell group (177), with statistical significance in three groups (P=0.000, for between group comparisons). Among these, the TAMs density and tumor invasion unit were two independent factors to impact on GC prognosis, while tumor neo-vessels density itself was not an independent factor for overall survival (OS). ROC analysis also demonstrated that tumor invasion unit had the largest area under the curve (63.2%), suggesting that tumor invasion unit had bigger predicting power than TAMs for prognosis prediction.Conclusion The co-evolution between cancer cells and tumor microenvironment played crucial roles in cancer progression. The characteristics acquired from key stromal components could be regarded as independent prognostic factors as well as promote the prognosis prediction and guide clinical treatment strategy. |
PDF Full Text Request