Spread by Lynaph or Blood



Micrometastasis Formation ^ ^ Stroma Activation

Continued Local Growth

Metastasis Growth

Figure 9.1 Steps of invasion and metastasis

(3) The tumor continues its growth into the connective tissue. This is one of the more variable steps. Some carcinomas continue to grow as solid, coherent masses compressing the neighboring connective tissue or develop processes that spread into it, breaking up the extracellular matrix (ECM). From other carcinomas, small groups of cells or single cells split off and migrate into the underlying tissue, sometimes in an 'indian file' pattern and sometimes as adherent clusters. These migrations, even by single tumors cells, involve remodeling of the ECM.

(4) In fact, the invasion of stroma by tumor cells is not as one-sided as it may appear. It is accompanied by altered gene activity in stromal cells, with some changes promoting and others inhibiting invasion. The type of stromal reaction may be one of the most important factors determining the ability of a tumor to metastasize. Often, cells of the immune system are attracted, by signals emanating from the stroma and from tumor cells. This can result in pronounced inflammation. Like the stromal reaction, the effect of inflammation is ambiguous; it may impede or promote invasion.

(5) A critical step in invasion is reached when the growing or migrating tumor cells encounter blood or lymph vessels and invade them. Like the previous steps, this can occur by a tumor mass growing through the vessel wall into the lumen or by single tumor cells squeezing through the vessel lining. By this 'intravasation' step tumor cells gain access to the circulation and can reach distal organs by 'lymphogenic' or 'hematogenic' routes. Gaining access to blood and lymph vessels is also not necessarily a one-directional process. Since many tumors induce angiogenesis, capillaries sprout from blood and lymph vessels into the direction of the tumor mass. Since these are often leakier than normal vessels, they may offer easier access to the circulation.

(6) Independent of whether metastasis occurs, invasion may continue into further layers of the organ from which the carcinoma arises, through a tissue capsule, into surrounding adipose tissue and into neighboring organs. An important spreading route for some cancers, e.g. of the kidney, liver, ovary and pancreas, is through the lumen of the retroperitoneum or the peritoneum ('transcoelomic metastasis').

(7) When tumor cells have entered into lymph vessels, they are transported to the filtering system of the local lymph nodes, where some may survive and start lymph node metastases. Cells from these metastases may eventually penetrate towards the main lymph vessels and eventually enter the blood by this route. Tumor cells or debris and signal molecules from tumor and stromal cells transported to the lymph nodes influence the immune reaction towards the primary tumor.

(8) Tumor cells having entered into blood vessels can theoretically spread to any part of the body. However, they are larger than normal blood cells and are not well adapted for survival in a moving liquid11. Survival in the blood to reach distant tissues may be limiting for metastasis.

(9) To form metastases, carcinoma cells must exit from the circulation by 'extravasation'. Most often, this appears to take place in organs with microcapillary systems, such as the liver, the lung, the kidney, and bone. Because of their size, carcinoma cells (and certainly cell clusters) get stuck in capillaries. This is not sufficient, however, to establish micrometastases.

(10) In the new tissue environment, carcinoma cells have to reattach to the matrix, survive, sometimes for extended periods, and eventually start to expand into micrometastases, which again can lie dormant for many years. After all the complicated previous steps, it may be surprising to learn that this step is by many considered the most critical, i.e. least efficient step in metastasis formation.

(12) The final step in metastasis is the expansion of micrometastases to actively growing tumors. This requires establishment of a sufficient nutrient supply and interaction with a different type of stroma, often including once more induction of angiogenesis and further local invasion.

Although invasion and metastasis are such important processes in the course of cancer progression, they are incompletely understood. This is largely owed to their

11 In fact, carcinoma cells are often observed in blood as small cells clusters.

complexity, since at almost every step complex interactions between different cell types and extracellular tissue components are involved. In addition, the early steps of metastasis in humans can rarely be observed. Moreover, metastases specimens are not regularly available for investigation, particularly from carcinomas, since they are rarely treated by surgery. Therefore, much of our knowledge on this matter is inferred from experimental animal models, which are by themselves complex enough. At this stage, therefore, many individual factors involved in invasion and metastasis have been identified and selected interactions have been pinpointed (see below). However, there are considerable deficits in understanding the relative importance of this factors and how they interact with each other.

one particular important issue is what drives the overall process of invasion and metastasis. Two alternative ideas are entertained. one hypothesis maintains that tumor cells acquire one property after another as they proceed through the steps outlined above. At each step (or at least at many), the best adapted tumor cells are selected from the numerous variants created by inherent genomic instability. The alternative hypothesis holds that the 'invasion and metastasis' program is an inherent property of certain cancers that is expressed very early on in their development. This second hypothesis would predict that primary cancers are more similar to their metastases than to each other in their genetic alterations and gene expression patterns, and that primary cancers which metastasize show more similar alterations to one another than to those that do not. This has indeed been observed in some investigations using expression profiling methods, e.g. in breast cancers (^■18.5). Current opinion therefore leans towards the second hypothesis. Interestingly, and perhaps not unexpectedly, the most striking differences between metastatic and non-metastatic cancers were found in the gene expression pattern of the stromal rather than the tumor cells. The distinction between these two hypotheses is important for cancer diagnosis (^21.4). The prognosis of a cancer can only be determined from a sample from the primary site, if the ability to metastasize successfully is reflected in molecular parameters of the primary tumor. It would be difficult, if the primary cancer was a mixture of cells with different abilities to metastasize and/or if this ability developed gradually.

0 0

Post a comment