James Butera, Martin Malachovsky, Ritesh Rathore, Howard Safran

Departments of Medicine, Brown University, Providence Rhode Island

Received 6/12/98 Accepted 8/6/98

3. MATRIX METALLOPROTEINASE INHIBITORS

Matrix metalloproteinases (MMP’s) are a family of Zn²⁺ dependent endopeptidases with a broad spectrum of proteolytic activity for several components of the extracellular matrix(11). Tumor cells secrete MMP’s which destroy basement membranes and local connective tissue, allowing tumor cells to gain access to the lymphatic or blood circulation (12). Once established at a secondary site, tumor cells continues to secrete MMP’s that degrade connective tissue enhancing local growth (13). MMP’s also appear to promote the growth of new blood vessels that nourish metastatic deposits.

Though MMP’s are different in structure, molecular weight and substrate specificity, they all contain highly homologous zinc-binding active sites, hemopexin-like domains, and cleavable NH2-terminal sequences the removal of which results in activation of the enzymes (14). MMP’s are involved not only in pathologic tissue destruction by tumor cells. Their presence was documented during mouse embryogenesis (15). Collagenase activity was reported as important factor during wound healing and tissue remodeling (16). Several matrix metalloproteinases were identified to be secreted by rheumatoid synovial cells and thus contributing to joint destruction in rheumatoid arthritis (17).

In normal human tissues, MMP’s are in an inactive proenzyme form and their activity is regulated by its activators and inhibitors (18). A ubiquitous glycoprotein, TIMP, is considered to be a major inhibitor of metalloproteinase activity in tissues. This inhibitor is secreted by many cells in culture including fibroblasts , endothelial cells, chondrocytes and vascular smooth muscle cells . It is also present in bone, cartilage and amniotic fluid. TIMP inhibits MMP by forming irreversible 1:1 stochiometric complex with the active enzyme (18). During tumor invasion the balance between MMP’s and their inhibitors is broken and number of MMP’s exceeds TIMP’s which contributes to the invasion and degradation of extracellular matrix (19-22).

Early studies on human cancer cell lines and animal models have demonstrated that inhibition of MMP’s inhibit the growth and spread of primary tumors and promote the formation of stroma causing encapsulation of the tumor. Some additional beneficial effects on inhibition of angiogenesis were demonstrated as well (23-27).

Many solid tumors, including pancreatic cancer, express high levels of matrix metalloproteinases (28-30). Synthetic inhibitors of MMP’s are therefore being developed to counteract the destructive and invasive nature of these enzymes (31,32).

Marimastat was one of the first MMPI’s to show significant oral bio-availability to enter clinical trials. It is a synthetic MMPI that mimics the substrate of the matrix metalloproteinases. This allows it to fit tightly in the active site of the enzyme. Its hydroxamate group binds to the zinc atom in the active site resulting in potent but reversible inhibition of the metalloproteinases MMP-1 , MMP-2, MMP-3 and MMP-9.

Initial human trials occurred in normal volunteers, underscoring the fact that marimastat is not a cytotoxic agent. These studies did not indicate untoward toxicity, although some alteration of liver enzymes were noted. Following this, more than 1000 patients with metastatic ovarian, colorectal, pancreatic, prostatic, head and neck, breast, gastric, lung, and melanoma (33-38), have been treated on phase I/II trials to try to identify a biologically effective dose and to further to evaluate the safety profile and pharmacokinetics.

A phase I/II study of marimastat in patients with advanced, non-resectable pancreatic carcinoma has been reported (38). Patients received marimastat on a BID oral schedule for 28 days. One patient developed acute rash, fever, chills and muscle pains after marimastat. Otherwise there have been no significant drug-related toxicities. Seven of 19 patients demonstrated a reduction in the rate of rise of CA 19-9. Marimastat is also being tested by the Brown University Oncology Group in patients with locally advanced pancreatic cancer with stable or responding disease after paclitaxel and radiation. No significant toxicity from marimastat has been been seen thus far in this study.

Since marimastat is not a cytotoxic agent, traditional endpoints of objective tumor response are anticipated. The evaluation of disease stabilization in a uncontrolled setting is hard to interpret. Carefully designed phase III trials are needed to determine the effectiveness of marimastat. Therefore a multicenter randomized placebo controlled trial is underway in the United States to evaluate whether marimastat can improve disease free and overall survival in patients with resected pancreatic cancer.