The Future of Melanoma: Q&A with Martin McMahon, PhD
Martin McMahon, PhD, is the Cumming-Presidential Professor of Cancer Biology and Senior Director for Preclinical Translation research at Huntsman Cancer Institute (HCI) of the University of Utah in Salt Lake City. Prior to working at HCI, he was the Efim Guzik Distinguished Professor of Cancer Biology, Co-Leader of the Experimental Therapeutics program, and Director for Professional Education at the Helen Diller Family Comprehensive Cancer Center at the University of California, San Francisco.
For more than a decade, Dr McMahon has studied the mechanisms of metastatic melanoma, as well as lung, pancreatic and thyroid cancer, with a focus on the RAF protein kinases. He spoke with The Dermatologist about what he has learned as a melanoma researcher, the biggest breakthroughs of the past decade, and the future of melanoma research and treatment.
Q: How did you become interested in melanoma research and what does your laboratory primarily study?
A: I came into this field because the genes I worked on, most notably BRAF, turned out to be particularly relevant to the development of melanoma.
My research focuses on how normal melanocytes are converted to melanoma cells and, once that has happened, how to kill those cells as a way of trying to understand potential new therapeutic interventions. We use a variety of research tools to do these experiments, including cells that we culture in petri dishes, genetically engineered mouse models of melanoma, and patient-derived xenografts, which are tumor fragments taken from a patient and implanted and serially propagated in immunocompromised mice. We believe these patient-derived xenografts are a complementary model system to the other systems that we use for studying melanoma.
About 16 years ago, melanoma, which was previously thought to be a monolithic disease, was revealed to be a series of related diseases that have a unified pathology but are driven by different types of genetic abnormalities. My lab is interested in the RAF kinase family, and 1 of those kinases, BRAF, was identified as being altered and turned on inappropriately in about 50% of melanomas. We have focused on the BRAF subset of melanoma for about 15 years now.
Additionally, we are interested in how to avoid resistance to drugs that target BRAF signaling. In 2011, the FDA approved vemurafenib (Zelboraf), which is a specific inhibitor of the mutant form of BRAF that is found in melanomas. Patients given that drug can show dramatic responses to vemurafenib treatment, but more often than not those patients will develop recurrent and drug-resistant disease. We would like to understand in greater depth what the mechanisms of resistance are, many of which have been elucidated by numerous other laboratories. Additionally, we are trying to understand how we can prevent the onset of drug resistance in experimental models in the lab and, ultimately, in clinical trials.
For example, we published a paper in 2013 with Darren Stuart, PhD, and collaborators from Novartis, that assessed whether changing the dosing of therapies like vemurafenib could delay the onset of drug resistance.1 There is an ongoing clinical trial, which grew out of that work, that is testing whether patients receive as much or even more benefit from intermittent dosing of BRAF inhibitors as opposed to continuous dosing. The results of that trial will probably be available in 2020 or 2021, depending on how long it takes to accrue data.
We are also interested in a subset of melanomas called NRAS that are driven by that oncogene. NRAS-driven tumors are not particularly sensitive to pathway-targeted therapies. We have been trying to determine whether there are combination pathway targeted therapies that are effective for treating NRAS mutant melanoma, and currently have a paper under review that uncovered an interesting strategy for targeting those specific melanomas.
Q: What have been some advances in melanoma research?
A: It is an incredibly exciting time to be working in melanoma considering that, as recently as 10 years ago, melanoma patients with advanced disease almost invariably died, and clinical trials were uniformly negative. Within the past decade, we have gone from having almost no effective therapies for melanoma to having many. The challenge now is choosing the right strategy and determining the most effective order for those therapies, especially now that genetic subsets can help determine which path you should take.
In 2011, the FDA approved 2 separate drugs for melanoma treatment: vemurafenib and ipilimumab (Yervoy), a drug that stimulates the immune system to kill melanoma cells. This remarkable confluence of drug approvals represented many years of work and suggested that we were actually starting to make some progress against a disease that was previously utterly recalcitrant and almost utterly resistant to any type of therapy.
Now, we have about 3 classes of melanoma therapies: (1) Pathway targeted inhibitors; (2) Checkpoint blockade immunotherapies and; (3) A virus, called T-VEC, which has been shown to infect melanoma cells and stimulate the immune system to recognize and kill them. What we would like to be able to do is understand if there are ways to combine these modalities together to treat melanoma. For example, if a patient has BRAF mutated melanoma, he or she could receive either the virus or immunotherapy or pathway targeted therapies. The question is: Is there a rational way to administer some combination of these therapies that would provide the patient with the benefits of each? I think this is a major research gap, and a clinical gap as well, that we need to think about and explore in experimental model systems in the laboratory.
The ability to combine these agents in a rational way to come up with predictions that could be tested in clinical trials would really open up the possibility of curing more patients with melanoma, because it is clear now that immunotherapy can sometimes be curative.
Now, patients with advanced melanoma can be cured, which is kind of a weird thing to say because as recently as 5 years ago no one would have dared say the word “cure” in the context of the phrase “advanced melanoma.” With the ability of immunotherapy to mobilize the immune system, we are providing long-term and durable responses. After you get past about 5 years, doctors are willing to say the words “potentially cured.”
Q: What are some of the gaps in melanoma research?
A: There is an interest in chemoprevention with agents such as aspirin. There is some suggestion that aspirin might have chemopreventive effects for individuals with a high risk for melanoma, who have skin damage, and/or are from a family with a high risk for melanoma.
We now understand that there are about 4 subclassifications of cutaneous melanoma: BRAF, NRAS, NF1, and triple wildtype. These 4 classes will be treated very differently due to the fact that we have different agents. Some of these subtypes, especially the BRAF subset, are remarkably sensitive to inhibitors that block that signaling pathway. However, new and innovative strategies to prevent the onset of drug resistance are starting to come to the foreground. There are clinical trials being conducted on the efficacy of combination therapies that might really work and be able to treat individuals with drug-resistant melanoma.
While we have made great strides in developing treatments for melanoma, we still do not fully understand the right way to treat melanoma patients with the various forms of immunotherapy available. While we know that the response rates and the long-term remission in patients is maybe around 20% to 25%, we cannot identify upfront which patients are going to benefit from the immunotherapy. This remains a major challenge.
Unlike BRAF inhibitors, where we know how to stratify patients based on whether they do or do not have the BRAF mutation, we do not yet have well validated predictive biomarkers that allow us to stratify patients for treatment with immunotherapy. Among patients receiving immunotherapy, there is a fraction that will not receive any benefit, some who will benefit from it, and a percentage who will not only benefit from it but be cured. At the moment, we cannot tell these 3 different subsets of patients apart.
Predictive biomarkers for determining who will benefit from immunotherapy is a really important area for future research. It will help identify patients who will benefit from immunotherapy and cut down the administration of immunotherapy to those who get no benefit. These immunotherapies can have substantial adverse effects and are very expensive. If we can determine who will and will not benefit from immunotherapy, we can provide patients who will not benefit with other types of therapies or suggest that they participate in clinical trials of new melanoma therapies that are not yet FDA-approved (Table).
Q. Have there been any advances in identifying biomarkers for predicting disease progression?
A: There has been a long-standing interest in trying to understand if there are biomarkers in early stage disease that would predict future disease recurrence accurately. Most patients who are diagnosed with early-stage melanoma will be cured by surgical intervention, but a certain percentage, maybe about 10% of patients with early-stage melanoma, will develop recurrent disease. These patients will come back to the clinic 1, 5, 10, and sometimes 20 years later with aggressive, often metastatic melanoma.
This was something I encountered in my personal life. Shortly after my family and I arrived in Salt Lake City, my wife and I became friends with our babysitter and her stepmother, who had been treated for melanoma 20 years previously. She recently had a breast examine that identified a lump in her breast, which turned out to be recurrent melanoma that was most likely from the melanoma she had removed 20 years ago. Although she received the best treatment that medicine can offer, including immunotherapy and BRAF-targeted therapy, sadly she died very shortly after we got to know her.
This is a major problem in the United States and worldwide. I think the rough estimate now is maybe 90,000 cases of melanoma that are treated just by surgical excision in their early stage and so dermatologists quite rightly believe that in many cases the patients have been cured. However, a small but certain percentage of those patients will eventually recur. It would be great if we could understand what the predictors of recurrence are in order to follow those patients on a more regular basis to see if we can catch the disease before it comes back in a widely metastatic and aggressive state.
This is a very important area for future research, especially for practicing dermatologists who would be able to recognize the risk factors for melanoma progression. Currently, most of the risk factors we assess are relatively simple, such as how thick is the lesion, is it ulcerated, and does it have large numbers of mitotic figures.
Although there are companies marketing genetic tests that they believe can predict future disease recurrence, this remains somewhat controversial in the field, and it will take years of additional testing to determine how well these genetic tests work and whether they can accurately predict disease recurrence. In my opinion, there is nothing at the moment that really qualifies as an accurate and specific genetic test that would predict future recurrence of early-stage cutaneous melanoma.
Article continues on page 2