Recent advances in our understanding of the molecular mechanisms of pancreatic cancer progression are resulting in novel approaches to therapy. The matrix metalloproteinases (MMPs) have emerged as promising new targets for therapeutic intervention strategies. More than 20 MMPs have been identified, which can be subcategorized according to preferential substrate.^[1]

MMPs are frequently overexpressed in human tumors and generally correlate with cancer stage. Animal models have shown that matrix metalloproteinase inhibitors (MMPIs) can impair the growth of solid tumors, inhibit metastatic spread, and block tumor neovascularization. Activity is enhanced when MMPIs are used in combination with cytotoxic chemotherapy.^[2] A number of different MMPIs have been developed for clinical use. They can be loosely grouped by target specificity. Marimastat is a broad-spectrum synthetic MMPI that has no direct cytotoxic effects on cultured cancer cells but has slowed tumor growth and metastases in a variety of preclinical models.^[3]

Bramhall and colleagues describe the results of a multicenter trial involving patients with unre-sectable pancreatic cancer who were randomly assigned to 1 of 3 dose levels (5, 10, or 25 mg bid) of marimastat or chemotherapy with gemcitabine (1000 mg/m²).^[3] There was no significant difference in overall survival between the gemcitabine and marimastat groups in the primary analysis, although median survival seemed to be superior in the group treated with gemcitabine. While the survival in the high-dose marimastat group seemed inferior to gemcitabine at 6 months, the 1-year survival rates were nearly identical, suggesting the potential for delayed efficacy. Adjusting for baseline prognostic factors, the survival in patients treated with gemcitabine was significantly better than it was for patients treated with the 2 lower dose levels of marimastat. Patients with metastatic pancreatic cancer seemed to fare particularly poorly on marimastat, suggesting that its efficacy is stage-specific. Interestingly, the greater myelosuppressive effect of gemcitabine did not compromise quality of life relative to marimastat.

The lack of frank tumor regression with MMPIs in animals has fundamental implications in terms of clinical trial design. New paradigms for clinical development are needed to assess the efficacy of agents like marimastat that slow, but do not reverse, tumor growth.^[1] In Bramhall and colleagues' trial, patients with progressive disease were removed from the study. However, given the potential for delayed activity, those patients who were treated with marimastat may have faired better if they continued to participate in the study beyond disease progression. Alternatively, advanced disease may not be the best setting in which to test marimastat as a single agent. Focusing on the adjuvant setting or combinations with cytotoxic chemotherapy may be more fruitful.^[1]

Defining the optimal dose of cytostatic agents presents another important challenge. Unlike cytotoxic chemotherapy, these agents may not have a maximum tolerated dose (MTD) or the MTD may not correlate with biologic activity. A dose response with marimastat in terms of both survival and toxicity was suggested in this study. The trough plasma levels of marimastat were related to dose and exceeded the 50% inhibitory concentration within the tumor. Ideally, the optimum biologically active dose of MMPIs should be defined by biologic end points in vivo. Recent advances in imaging techniques may enhance our ability to track drug activity noninvasively. One study group has successfully used an activated probe to monitor MMPI activity in intact tumors in nude mice.^[3]

Another major challenge related to the use of MMPIs is defining the molecular target. In recent years, the number of MMP types has grown to more than 20, so the specificity and selectivity of many of the MMPIs in clinical development may not be fully defined. Different tumors have unique MMP activation and expression profiles that may be cell-type- and stage-specific. Any given MMP may promote tumor development at one stage while inhibiting certain steps of tumor progression in another stage.^[4]

MMP expression in individual tumors needs to be carefully defined, and the role of MMPs with regard to tumor behavior and prognosis needs to be delineated. Although MMPs are best known for their capacity to cleave extracellular matrix proteins, recent evidence indicates that MMPs have other important cellular functions.^[1] Without fully understanding the role of target MMPs in tumor progression, cancer therapy based on MMP inhibition may not succeed.^[4]

In summary, it would be premature to say that marimastat is equivalent to gemcitabine in treating pancreatic cancer. Expansion of the findings of this trial should be pursued; however, future studies should be designed with end points consistent with the expected clinical effect. An assessment of marimastat activity in the adjuvant setting and/or in combination with chemotherapy may be most prudent.^[5]

References for:

[Abstr Hematol Oncol 5(2):5-6, 2002. © 2002 Cliggott Publishing, Division of SCP Communications ]

1. Hidalgo M, Eckhardt SG. Development of matrix metalloproteinase inhibitors in cancer therapy. J Natl Cancer Inst. 2001;93:178-193.

2. Nelson A, Fingleton B, Rothenberg M, Matrisian L. Matrix metalloproteinases: biologic activity and clinical implications. J Clin Oncol. 2000;18:1135-1149.

3. Bremer C, Tung CH, Weissleder R. In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat Med. 2001;7:743-748.

4. Chang C, Werb Z. The many faces of metalloproteases: cell growth, invasion, angiogenesis and metastasis. Trends Cell Biol. 2001;11(suppl):S37-S43.

5. Hess K, Abbruzzese J. Matrix metalloproteinase inhibition of pancreatic cancer: matching mechanism of action to clinical trial design. J Clin Oncol. 2001;19:3445-3446.