Matrix dissolution is a crucial step during tumor progression that converts a premalignant cell to an overtly malignant one. Main players in this step are the various matrix metalloproteinases (MMPs), which differ in substrate specificity and tissue distribution, and thereby also differ in presence and function during various stages of initial and systemic tumor spread. Accordingly, the inhibition of MMP synthesis and/or activity represents novel potential therapeutic strategies for the treatment of cancer patients. Considerable work has already been carried out on synthetic inhibitors of MMP activity, but with little or even adverse effects in recent clinical studies. The reasons may be inappropriate patient populations in too advanced tumor stages, or inappropriate enzymes as targets for inhibition. Upregulation of endogenous tissue inhibitors of MMP (TIMPs) also provided ambiguous results, since TIMPs possess biologic functions in addition to MMP inhibition, for example, TIMP-2 is a main player in the MMP-2 activation cascade. This may explain, at least in part, the adverse effects of TIMP application/upregulation. Other strategies have been sought in order to overcome these problems. These include the downregulation of MMP transcription by cytokines. However, the effects of cytokines (other than MMP inhibition) may also limit the use of this approach. Finally, empiric evidence for control and modulation of MMP transcription and/or activation by several naturally occurring substances, such as flavonoids, green tea polyphenols and curcumin, represent novel options for the control of MMP activity even in early tumor stages. Additionally, these substances have little or no toxic side effects and good bioavailability, and therefore their continuing analysis provides intriguing insight into tumor pathophysiology and possibly new therapeutic options.