Cancer Diagnosis and Prognosis

Despite the fact that most mutations do occur randomly as tumors progress to the metastatic state, some mutations are indeed associated with a particular stage in the metastatic process. For example, changes in regulation of cell proliferation or cell death often are among the earliest changes found in new tumors, long before the tumors become metastatic. Alterations in cell-cell adhesion, and protease production often are seen after the changes in proliferation but before metastasis is observed because they are required for cell invasion across the basement membrane and into the tissue stroma. Other mutations may be seen later in tumor progression as the cells are becoming or have become capable of metastasis.

Could the distinction in the time of appearance of a particular mutation be useful clinically? We believe that it is. For example, if we wish to know if a particular patient has or does not have cancer, we should use a marker that comes up very early and is expressed continuously by the tumor. In this way, all patients with that cancer will be detected by screening for that marker. In contrast, a marker that comes up when a tumor is converting from non-metastatic to metastatic may be useful in tumor prognosis. A patient without evidence of that marker may have a tumor that is still confined to the primary site and can be eliminated by radiation or surgery. In contrast, a tumor that produces a metastatic marker may have already produced micrometastases at a distant site and require systemic treatment with chemotherapy or anti-angiogenic agents. Markers produced by aggressive tumor cells may also be used to immunize patients (tumor vaccines) and use the body's immune system to attack the aggressive tumor cells.

In our own lab, we have found that certain molecules that regularly appear later in metastatic progression can be used to predict the course of prostate cancer in patients. The PSA test, which is used to diagnose prostate cancer, is not as useful for telling whether a patient will go on to develop metastatic disease. In contrast, we have developed a panel of markers that appear to be able to predict which tumors are likely to go on to metastasize or have likely already produced metastatic colonies. The prototype for this type of molecule is thymosin beta-15, a molecule that stimulates cell migration and promotes metastasis in prostate cancer cells. We and others have now shown that tumors in which thymosin beta-15 cannot be detected are unlikely to develop metastases and may not warrant aggressive treatment. In contrast, tumors expressing thymosin beta-15 are more likely to have disseminated metastases and are candidates for aggressive systemic therapy. (A more complete description of this molecule can be found in the section of this web site devoted to prostate cancer.) A large number of cancer biomarkers are currently being developed in our laboratory and by others in our program (see Dr. Marsha Moses). Such markers can be used for several purposes including: 1) detection of early cancer; 2) prediction of tumor cause and outcomes; 3) detection of recurrent cancer; and 4) determination of appropriate therapy.

Beyond developing cancer biomarkers, a second goal of our laboratory is the development of new therapies that specifically target metastatic tumors. Given our expertise in the understanding of the process of metastasis, along with new tools for the screening of drug libraries, we are beginning to identify therapies that selectively target metastatic tumors.