20 October, 2001

Micrometastasis is tumor invasion in a tumor-host microenvironment, where stroma and tumour cells exchange enzymes and cytokines that modify the local extracellular matrix, stimulate migration, and promote proliferation. Malignant tumour cells recruit vasculature and stroma through production and secretion of stimulatory growth factors and cytokines, and local host stroma are activated for tumor cells. Normal cells usually remain confined to the territory where they are supposed to be because they are held in check by interaction with neighbouring cells and with the surrounding extracellular matrix (ECM). But metastatic tumor cells are somehow resistant to these regulatory signals, and additionally there are indeed specific molecules identified to positively promote metastatic progression. During metastasis, tumour cells leave the stroma, enter nearby lymphatic and blood vascular channels, circulate, attach to endothelium in the target organ and are stimulated to grow as colonies inside the vessel. Steps in this pathological process can be viewed as targets for the development of novel cancer therapeutics. Conceptually, Loitta LA et al (2001: uid=11357145) suggest stromal therapy as a new strategy for cancer intervention. Stromal therapy here may be referred to as cancer prevention and intervention by addressing to early molecular signals that participate in tumourﾐhost cross-talk. There are already a large number of inhibitors and modulators being developed to intervene several distinct steps in the cross-talk.

In a short review, Loitta LA et al (2001: uid=11357145) describe enzymes and factors involved in the first kind of local invasion steps of metastatic tumors where ECM and associated proteins are degraded. These may be classified as 1) the matrix metalloproteinases (MMPs), a family of secreted and membrane-anchored proteinases, 2) the adamalysin-related membrane proteinases, 3) the bone morphogenetic protein-1-type metalloproteinases, and 4) tissue serine proteinases, including tissue plasminogen activator, urokinase, thrombin and plasmin. It should be pointed out that in this hostﾐtumour interdependence, most of the enzymes complexed at the invasion front are contributed by host cells, not by the invading tumour cells. These enzyme classes are confined to the point of invasion by binding them to specific receptors, integrins, adjacent ECM molecules and other heterodimeric complexes. The direction of tumour-cell invasion and migration can also be influenced by local chemoattractants that include cytokines and growth factors. As for leukocytes and stem cells to home into the specific organs by chemokine mechanisms, tumor cells appear to use similar mechanisms to direct metastatic organ preference. For example, breast cancer metastasizes preferentially to the regional lymph nodes, bone marrow, lung and liver. Muller A et al (2001: uid=11242036) have shown that chemokine receptors CXCR4 and CCR7 were highly expressed in human breast cancer cells, and their respective ligands CXCL12/SDF-1alpha and CCL21/6Ckine exhibited peak levels of expression in organs preferred for breast cancer metastasis. In vivo, in model immunodeficient mice, metastasis to CXCL12-rich human lung tissue was blocked by the treatment with a neutralizing anti-human CXCR4 monoclonal antibody. In vitro, CXCL12 stimulated breast cancer cells to sent out extensions (pseudopodia) that migrated in a directed manner, and penetrated barriers imposed by extracellular matrix. It appears that breast cancer metastasizes to the lung by a receptor/ligand mechanism, namely the chemokine receptor CXCR4 on the cancer cell interacts with the ligand CXCL12 on the preferred organ tissues. A similar receptor/ligand mechanism may also be true for malignant melanoma that express high levels of CCR10, in addition to CXCR4 and CCR7. These findings naturally suggest that small-molecular antagonists of chemokine receptors may be useful in treating cancer patients.

Once metastasis is established and tumor colonies grow inside the vascular vessel of target organ, the cancer intervention may be a totally different story than the block of metastasis process. Therefore it would be important to diagnose metastasis-prone tumors while still localized and prior to the spread in order to achieve best kind of treatment and high probability of survival of the patient. Tumorigenicity here may refer to the ability of cells to proliferate continuously in the absence of persistent stimulation, while tumor progression may be defined as the evolution of already tumorigenic cells toward enhanced self-growing states (malignancy). Perhaps metastasis, namely the ability of malignant cells to invade through basement membrane, is the ultimate hallmark and state of malignancy. As described above and as in Welch DR et al (1999: uid=10451432), a large number of possible markers have been reported to characterize metastatic tumors and relevant processes. They may be grouped as 1) those that are more highly expressed in metastatic tumors than in their nonmetastatic counterparts, 2) those functional and molecular changes that correlate with metastasis, and 3) those that positively regulate metastatic processes, and 4) those that suppress metastatic processes.

In fact, malignant tumors contain few preexisting metastatic subpopulations. Metastasis is a highly selective process, and it is estimated that less than 0.1% of cells entering the blood stream successfully form clinically detectable metastatic lesions (Fidler IJ, 1991: uid=1755831). Only a small percentage of primary tumor cells are expected to express markers of metastatic cells, and it would be rather difficult here to pin-point the essential markers. In practice, it would be preferable to identify those markers that positively regulate and/or suppress metastatic processes, whatever the mechanism may be. As metastasis involves multisteps, the block of one of them might eventually prevent its completion. In an effort toward early diagnosis and intervention, Shih JY et al (2001: uid=11562390) describe collapsin response mediator protein-1 (CRMP-1) gene that is potentially an invasion suppressor. From the 9600 genes screened in a panel of lung cancer cell lines, CRMP-1 was selected and transfected into lung cancer cell lines. Highly invasive cells transfected with CRMP-1 had lower invasive activity than untransfected cells. The expression of CRMP-1 messenger RNA (mRNA) in 80 tumor samples (non-small-cell lung cancer) was significantly lower statistically than in adjacent normal tissue. When the median value of CRMP-1 mRNA was used to classify patients into high-expression or low-expression groups, low-expression patients were more likely to have advanced disease and lymph node metastasis. The median duration to postoperative recurrence was also longer in high-expression patients (30.5 months) than in low-expression patients (15.9 months), and high-expression patients had a significantly longer survival statistically than low-expression patients. Thus CRMP-1 might be a marker for invasive phenotype of malignant cells.

The findings described above suggest potential targets for developing drugs against metastasis. It is annoying to realize that once malignant cells have already metastasized to ectopic sites and growing inside the vascular vessel, anti-metastatic drugs may not work anymore. There is not much information on the dynamic aspects of metastasized cells in the target organ. Are they more or less similar to the primary cancer cells? Are they somehow limited and/or in check in growth potential in the target organ environment and still vulnerable to anti-metastatic drugs? It should be seen how anti-metastatic drugs behave in clinical trials. Heparanase (an endo-beta-D-glucuronidase) that degrades heparan sulfate peptidoglycans in ECM is not discussed here but it emerges as a very promising target for developing anti-metastatic drugs. Please go to Heparanase: A New Target for Anti-metastatic Drugs for details in this series of News and Commentaries.