Melanoma is the rarest form of skin cancer, but also the deadliest one. Due to its high propensity to metastasize, by the time it is diagnosed, it often is incurable. So while melanoma accounts for less than 5 percent of skin cancer diagnoses, it is responsible for 75 percent to 80 percent of skin cancer deaths.
Two recent papers reported findings that could translate into advances against melanoma on both the prognostic and the therapeutic fronts. One of the papers also helped explain some earlier clinical failures. At least some members of a group of proteins named matrix metalloproteinases or MMPs, long believed to be oncogenes, appear to be tumor suppressors instead.
"For many years, MMPs were thought to promote metastasis," senior author Yardena Samuels told BioWorld Today. That idea began to be questioned as MMP inhibitors flopped in the clinic; the drugs were ineffective at best, and sped up cancer progression at worst. Some animal studies gave hints that matrix metalloproteinases might be acting as tumor suppressors. But Samuels said the new work, published in the March 30, 2009, online edition of Nature Genetics, "is the first study showing that it's true in human cancers."
In general, Samuels said, MMPs cut a "complex and diverse" group of substrates, which makes it challenging to determine their mechanism of action. But the idea that MMPs, which cut extracellular matrix proteins like collagen, could liberate cancer cells and enable them to invade distant sites and metastasize, certainly had logic going for it.
Logic, but not evidence. When Samuels, who is an investigator in the Cancer Genetics Branch of the National Human Genome Research Institute, and her colleagues studied about 80 blood and tumor samples from nearly 30 patients with aggressive melanoma, they found that roughly a quarter of them had mutations in eight different MMP genes. The largest number of mutations were found in two structurally similar enzymes, MMP-8 and MMP-27.
The authors concentrated their work on MMP-8,which already is known to be involved in skin cancer. Tellingly, mutations did not increase the activity of MMP-8, as would be expected of an oncogene. Instead, the mutated proteins were less able to cut collagen. Collagen is a major target of MMP-8, at least in cell culture, though Samuels pointed out that "we don't know what the physiological substrate is for MMP-8. . . . Collagen is a good substrate [in culture, but] whether it's the in vivo substrate, we don't know."
Her team is working on establishing the substrate, which could lead to therapeutic targets, though she noted that activators are more difficult to develop than inhibitors. "The optimal situation is if you find a mutation that activates a protein," Samuels said.
In the meantime, she added, the findings underscored that one size does not fit all in cancer therapies. "One really needs to evaluate [therapies] in a cancer-specific, and even individual-specific, manner," she said.
Such mutations also potentially could be used prognostically, though at this point, the samples are too small to know for sure, she said. Another paper out recently, however, does provide such a prognostic tool: In the April 1, 2009, issue of Clinical Cancer Research, scientists reported that a variant of the MDM2 gene appeared to raise younger women's risk of developing melanoma nearly fourfold, a greater increase than that conferred by clinical risk factors such as blistering sunburns and family history. The findings could be used to identify individuals at high risk of developing melanoma.
Scientifically, they also may help explain the sex distribution of melanoma. The variant allows MDM2 to bind estrogen more tightly; MDM2, in turn, regulates levels of the tumor suppressor p53, so the stronger activation of MDM2 is bad news for p53 levels and, consequently, tumor suppression. That role for estrogen helps explain why women are more likely to develop the disease than men in the younger than 40 crowd, while the ratio is about 1:1 in those ages 40 to 50, and for those older than 50, men have a greater risk.