[If you cannot see the flash video below, you can click here for a high quality mp4 video.]
Interviewees: Mohammed Kashani-Sabet,
Malignant melanoma is among the most aggressive and fatal of cancers, but it can be treated if caught in its early stages. Unfortunately, misdiagnoses are not uncommon. In fact, melanoma misdiagnosis is the second most common cause of cancer malpractice suits in the US, behind breast cancer.
We and our doctors should routinely check moles on our bodies and be aware of the ABCDE’s of melanoma. When your doctor removes a suspicious moles, it’s up to trained pathologists to examine the biopsy sample under the microscope and decide whether it is a malignancy or just a benign growth.
"Many of the cases that are biopsied every day, some of them are clear-cut moles and some of them are clear-cut melanomas, and we don’t need a lot of help to be able to say this is what they are," explains Mohammed Kashani-Sabet, director of the Melanoma Center at the University of California, San Francisco Helen Diller Family Comprehensive Cancer Center. "But there’s a subgroup—and we’re not sure exactly what percentage of them are like that– but there’s definitely a group that are very difficult to diagnose.
Also on ScienCentral
And you have situations where one pathologist will look at it and they’re not sure, and they might send it to two or three other pathologists. And some of them might come down on the benign side and others might come down on, ‘this is a malignancy… and really, until now, we haven’t had a way of being able to test anything else to help us figure out whether it’s really a mole or a melanoma."
Microscopes and Microarrays
Now Kashani-Sabet, Mehdi Nosrati and their colleagues have developed a molecular test that can highlight differences between malignant melanomas and harmless moles under the microscope.
The University of California has patented the test, and Kashani-Sabet has already started up a biotech company to work on commercializing it. He hopes it will be in widespread use in a year or two.
It’s based on their previous finding that used DNA microarrays to analyze the levels of activity of more than 40,000 different genes in various stages of melanoma, as well as in benign noles. That work revealed "about a thousand genes that… the levels were different in these genes in moles versus melanomas," Kashani-Sabet explains.
In their new study, published in the Proceedings of the National Academy of Sciences, "We picked five of these genes that would be at higher levels in melanomas than in the moles. And by looking at these five genes and combining their levels together, we were able to develop a test to be able to distinguich between a mole and a melanoma," Kashani-Sabet says.
To develop the test, they used a technique called antibody staining to highlight the proteins encoded by those genes in biopsy samples, allowing them to visualize the amounts of the proteins under the microscope. They also used a computer program that can score the amounts of staining in each sample.
"We looked at these five proteins initially in 500 moles and melanomas and found that these markers, the proteins, were accurate in diagnosing these cases in about 92 percent of the time," says Kashani-Sabet.
It turned out that in addition to the amounts of these proteins, the team also found a difference in their locations.
"We found that the levels of these proteins were clearly higher in melanomas than in moles, and that’s what we expected from our earlier study," Kashani-Sabet says. "But what was unexpected was that it wasn’t just how high the proteins were, but the pattern of expression was very different between moles and melanomas."And what we found was that, for melanomas, these proteins expressed them in a very uniform way from the top of the cancer to the bottom of the cancer."
Whereas for the mole, the proteins were present at the top, but they were usually lost or absent down at the bottom. So this different pattern of expression wound up being able to help us discriminate between the benign mole and the cancer," he says.
In melanomas, the five protein markers were found throughout the biopsy sample.
In moles, the five protein markers were mostly present at the surface.
Testing the Test
To validate the new test, the researchers tested it on a separate set of 24 skin biopsy samples known to have been misdiagnosed.
"We said, ‘how often would these misdiagnosed cases be reversed with our markers?’ and it turned out that in these cases our markers would have gotten them right 75 percent of the time," says Kashani-Sabet. "That would suggest that this test that we’ve developed could prevent about three quarters of the errors that could come up from looking at skin biopsy slides and specimens."
Of the 24 misdiagnosed cases studied, 18 were actually malignant melanomas that had initially gone undiagnosed.
This research was published in PNAS Early Online edition for the week of March 30 – April 3, 2009, and funded by the Herschel and Diana Zackheim Endowment Fund, the American Cancer Society and the National Institutes of Health.
Elsewhere on the Web:
Doctors Question Rise in Skin Cancer Biopsies—NPR News, May 2007