By Abby Van Voorhees, MD, August 01, 2012
In this month’s Acta Eruditorum column, Physician Editor Abby S. Van Voorhees, MD, talks with Pedram Gerami, MD, about his recent American Journal of Surgical Pathology article, “Distinctive Clinical and Histologic Features in Cutaneous Melanoma With Copy-number gains in 8q24.”
Dr. Van Voorhees: Let’s start by talking about the classification of melanoma by the World Health Organization. Can you review this for us? What are the limitations and problems with this system?
Dr. Gerami: The current WHO classification system subdivides melanoma into four basic categories: superficial spreading, nodular, acral, and lentigo maligna types of melanoma. This classification is mainly based on a combination of clinical and histologic features of the melanoma, which was helpful as an organization tool at that time.
But the problem is that it really only conveys a very limited amount of information as far as further prognosis and does not necessarily group melanomas into categories with similar genetic abberrations. For example, a melanoma that’s in an area of chronic sun damage that’s designated as a superficial spreading type of melanoma might have more in common, genetically, with a melanoma that’s called a lentigo maligna in an area with chronic sun damage than it would with a melanoma that’s called superficial spreading that’s in an area of non-chronic sun damage. So there’s limited genetic commonality, limited prognostic information, and hence limited theragnostic information conveyed by the classification system, since a lot of the therapies are based on targeting the genetic commonalities.
Dr. Van Voorhees: More recently, what have we learned about the molecular pathways in melanoma? Are there characteristic gains or losses of specific chromosomal loci?
Dr. Gerami: If you look at melanomas and organize them by how they relate to sun exposure, there are some commonalties in the genetic abnormalities that occur. There are a number of chromosomal loci that are recurrently involved, either lost or gained. These loci contain genes that are very important in the regulation of the cell cycle, determining whether melanoma cells are going to proliferate or not. A group of specific sites that are commonly gained or lost have been revealed in the last 10 years. For example, melanomas that occur on areas of chronic sun damage have a much higher incidence of gains of a chromosomal loci of 11q13, where cyclin D1 is, whereas gains of that area of the chromosome are much less common in melanomas that occur in areas that are not chronically exposed to the sun. [pagebreak]
Dr. Van Voorhees: What did your study show? What were the features in the MYC mutant tumors?
Dr. Gerami: Our study took a group of melanomas that had very aggressive behavior — that had metastasized and often resulted in death for the patients — and matched it with a control group of melanomas that had similar standard prognostic parameters that have been developed by the American Joint Committee on Cancer, including Breslow depth, ulceration status, and mitotic count. These are the standard, conventional prognostic markers used to determine how aggressive a melanoma is going to be. We matched the two groups up for these standard parameters.
Then, using fluorescence in situ hybridization targeting the chromosomal loci that are commonly gained or lost in melanoma, we performed an analysis on both groups to see if there were certain specific gains or losses that would be much more characteristic of the group that acted in an aggressive manner. What we found is that two specific chromosomal loci that house two very important genes involved in cell cycle growth played a critical role, or were very highly correlated to aggressive behavior. One of these was the MYC gene, which is located at 8q24, and the other was the cyclin D1 gene, located at 11q13. When we found a high level of amplification or copy-number gains at either of these two loci, it correlated very strongly with very aggressive behavior in these melanomas, even when we compared them to other melanomas that had very similar standard prognostic parameters.
Dr. Van Voorhees: Is there any indication that ethnicity played a role in melanoma aggressiveness?
Dr. Gerami: We didn’t have that information in the study. We did have the age, sex, and tumor site, but not the ethnicity. Some data in other studies has suggested that melanomas in the Hispanic population often present at a more advanced or aggressive stage, but it hasn’t been determined yet if that’s because of lack of screening or if they have this more aggressive subtype of melanoma.
Dr. Van Voorhees: Did you look at other mutations as well, such as the BRAF or NRAS mutation?
Dr. Gerami: The melanomas that have the copy-number gains in MYC tend to be melanomas that occur in areas of non-chronically sun-damaged skin. They also have a similar rate of BRAF mutation as other melanomas from non-chronically sun-damaged skin, so about 50 percent of the time they will have a BRAF mutation. [pagebreak]
Dr. Van Voorhees: Can you explain why there isn’t more concordance between MYC and BRAF?
Dr. Gerami: Both aberrations tend to occur in melanomas that are from non-chronically sun-damaged skin. One of the interesting things we found about the MYC-positive tumors is they have a very characteristic pattern. They occur in areas of non-chronically sun-damaged skin, and they are almost always amelanotic, so they look like pink bumps as opposed to what a lot of people typically think of as melanoma, a dark black or dark brown irregular macule. They also tend to more commonly have a more nodular and occasionally a primary dermal pattern.
So they have these three characteristic features — non-chronically sun-damaged skin, amelanotic, and nodular — and about 50 percent of the time they will have a BRAF mutation. The MYC mutation is likely a later step in the progression of melanoma. You can probably have a BRAF mutation earlier or an NRAS mutation or possibly another unknown earlier mutation and as long as you localize to melanomas on non-chronically sun-damaged skin there is the possibility that you will subsequently develop these copy-number gains in MYC. But it doesn’t have to come from a melanoma whose first mutation is BRAF or NRAS. The commonality with BRAF is that both are much more likely in non-chronically sun-damaged skin.
Dr. Van Voorhees: What makes you think the MYC mutation is happening later?
Dr. Gerami: When we see MYC-positive melanomas, they tend to be very aggressive. It’s very uncommon that we’ll see a melanoma that has MYC gains and when you look at it, it’s at the very earliest stages of a growth, either an in situ melanoma or the beginning stages. Usually it looks like a more advanced tumor. Whereas you can find the BRAF mutation both in the very earliest stages of melanoma and in more aggressive tumors, which suggests that it occurs in an earlier stage.
Dr. Van Voorhees: Do you anticipate that this will be a clinical tool that we can soon use to screen our melanoma cases to guide their management?
Dr. Gerami: I think it could be potentially very useful for a number of reasons. One is that if we identify these aberrations it provides us a lot of prognostic information beyond the standard prognostic parameters and can help us guide management, particularly deciding on imaging tests and which patients should be more aggressively imaged. There’s a lot of debate about how much imaging we should do on cancer patients because each time we image them we expose them to radiation. If you had something to better advise you as to which patients are most likely to have more aggressive disease that would be very useful.
The other thing is guiding therapy. The MYC gene encodes for a transcription factor which activates transcription of a number of key cell-cycle regulators activating cell-cycle proliferation downstream of the MAP kinase pathway. Therefore melanomas with BRAF activation and increased activity of MYC as a result of MYC amplification would likely continue to have continued uncontrolled proliferation despite inhibition of BRAF. Although more studies are needed to evaluate this, it is possible that by identifying chromosomal copy number changes such as MYC amplification, we may be able to identify melanomas which may show greater resistance to BRAF inhibition and may be good candidates for treatment with combination therapy, including BRAF inhibition with a second drug. [pagebreak]
Dr. Van Voorhees: Is there reason to believe MYC screening could be helpful?
Dr. Gerami: If you look at uveal melanomas, in the eye, the people treating them have not had very good luck using standard prognostic parameters like Breslow or ulceration status, and they have recently found that MYC amplification identifies a subgroup of uveal melanomas that are very aggressive. In fact FISH analysis targeting 8q24 is already being incorporated in the staging of uveal melanomas. By evaluating two chromosomal loci, 3p21 and 8q24, where the BAP1 and MYC gene are located, they can predict aggressiveness and likelihood of metastasis in uveal melanomas with a high degree of accuracy.
Dr. Van Voorhees: Are there drugs on the horizon that may help inhibit this mutation?
Dr. Gerami: There actually are some drugs in development for inhibiting MYC. It’s very hard because MYC is a transcription factor it encodes a protein that then activates a number of other genes that push cell-cycle growth. Inhibiting those types of proteins has always been very difficult. But one of the things we know is that MYC is involved in a number of other cancers; for instance, Burkitt’s lymphoma and a number of other hematologic malignancies have MYC involvement. There are a number of drugs currently in development that have been brought forward with the goal of treating some of the hematologic malignancies that have MYC aberrations. If those are successful there’s potential for them to be used to treat melanoma as well.
Dr. Gerami is associate professor of dermatology at Northwestern University’s Feinberg School of Medicine. His article was published in the American Journal of Surgical Pathology, Vol. 36, No. 2, p. 253264. doi: 10.1097/PAS.0b013e31823425cc.