For the longest time, using matrix metalloproteinase inhibitors to combat cancer has pretty much been one of those things that should work, but hasn't.

Matrix metalloproteinases are a group of enzymes involved in breaking down the extracellular membrane. The gelatinase subgroup consists of MMP-2 and MMP-9 and has been implicated in cancer metastasis. The gelatinases break down tissues that surround malignant tumors, and inhibiting them should prevent tissue breakdown and contain the malignancy, thus preventing metastasis.

Or that's the theory, anyway. In practice, British Biotech plc (now Vernalis plc), Celltech Group plc and a gaggle of big pharma companies have all discontinued clinical trials with broad-spectrum MMP inhibitors because they were either ineffective or, in some cases, performed worse than placebo.

However, given that the basic theory is sound, researchers are not yet ready to give up on the gelatinases. And new approaches are yielding results that finally might allow the reaction to gelatinase inhibitors to progress from underwhelming to, perhaps, something worth pursuing.

"Because a few clinical trials did not produce effective results, we should not consider that the battle has been lost," said Rafael Fridman, professor of pathology at Wayne State University School of Medicine. "All the studies show that [matrix metalloproteinase activity] is an important mechanism of metastasis."

Broad-spectrum inhibitors need to be administered at fairly low doses because their nonspecificity makes them toxic at higher ones; also, Fridman pointed out that they were tested in fairly late-stage cancers, in which they might no longer be effective.

More specific compounds might overcome some of the problems of broad-spectrum inhibitors, and in the May 2005 issue of Cancer Research, Fridman and his colleagues from the Technical University of Munich in Germany; Wayne State University in Detroit; and the University of Notre Dame in Indiana reported on such a compound. It affects MMP-9 and MMP-2 by what Fridman terms "suicide inhibition."

The compound, known as SB-3CT, works by binding "to the catalytic site of the gelatinase and is itself cleaved by the enzyme," Fridman explained. After it is cleaved, SB-3CT is modified in such a way that it remains bound to the gelatinase, thus inhibiting it for a much longer time period than a broad-spectrum inhibitor would.

In their article, the scientists pretreated mice with varying doses of SB-3CT before injecting them with a tumor cell line. Treatment reduced the number of liver metastases in all groups, with the highest dose being the most effective. That highest dose also reduced the size of metastases and lengthened the mean survival time, though mortality eventually did reach 100 percent in the treatment group. The authors nevertheless called the effect on survival "remarkable, considering that SB-3CT is a prototype inhibitor with poor solubility and the aggressive nature of this tumor model, which is well documented."

When given after metastases had formed, SB-3CT was ineffective in preventing the formation of metastases. Asked whether that concerned him, Fridman responded that "this is a very aggressive lymphoma. Possibly, [SB-3CT] will be effective when given after the cancer cells with a less aggressive tumor model. We will have to do the experiment."

SB-3CT's inhibition of MMP-9 affects cells in several ways. "It is not toxic to cells, so it doesn't work like conventional chemotherapy," he said. "Instead, it won't allow the tumors to spread to other sites." It also indirectly inhibits cell growth, by affecting the cleavage of proteins necessary to conduct it.

One mechanism of action of MMP-9 inhibitors that is well known, but probably irrelevant in the experiments, is angiogenesis.

"The liver is a very well vascularized tissue, so [anti-angiogenesis] is not an important factor in in this liver metastasis tumor model," Fridman said. "That doesn't mean the inhibitor doesn't inhibit angiogenesis - it does. But in this particular model, that is not likely to be its main mechanism of action."

The Cancer Research paper follows on the heels of another announcement, made in April at the AACR meeting in Anaheim, Calif., that Montreal-based Procyon BioPharma Inc. identified a receptor and a mechanism of action for its drug candidate PCK3145, another MMP-9 inhibitor. The peptide appears to work by binding to two cell-surface receptors, the laminin receptor and the vascular endothelial growth factor receptor.

"Laminins are a family of extracellular matrix proteins that promote the invasive characteristics of tumors," said Richard Béliveau, professor at the University of Quebec in Montreal and a member of Procyon's scientific advisory board, in a conference call. He has studied the drug's mechanism of action. "PCK3145 binds to the cell-surface laminin receptor and triggers a signaling cascade of events, resulting in the down-regulation of MMP-9 expression that was observed in the clinic."

In the call, Béliveau added that the drug also antagonized phosphorylation of both the vascular endothelial growth factor receptor and VEGF-induced erk phosphorylation.

"These latter observations established the role of PCK3145 as anti-angiogenic agent, because the VEGF receptor is a very important target for angiogenesis," he said.

A Phase IIa study conducted last year in the UK had shown PCK3145 to be safe and effective in normalizing plasma MMP-9 levels. Procyon has filed an investigational new drug application for PCK3145 to enter a Phase II North American study, and expects to begin the clinical trial, which will be spearheaded by the Memorial Sloan-Kettering Cancer Center in New York, by the end of 2005.