Using gene expression profiling to identify risk of melanoma metastasis
Clinical Applications
Dr. Schwarzenberger is the former physician editor of DermWorld. She interviews the author of a recent study each month.
By Kathryn Schwarzenberger, MD, June 3, 2019
In this month’s Clinical Applications column, Physician Editor Kathryn Schwarzenberger, MD, talks with Sancy Leachman, MD, PhD, about her recent JAAD article, “Identification of patients at risk of metastasis using a prognostic 31-gene expression profile in subpopulations of melanoma patients with favorable outcomes by standard criteria.”
Dr. Schwarzenberger: Can you address the recent AAD guidelines that do not advocate routine gene expression profiling (GEP) testing on melanoma patients?
Dr. Leachman: The AAD Guidelines Committee evaluated all the available literature to support all practices in the diagnosis and care of melanoma patients. This committee was comprised of experts in our field and they did not feel that the data currently available supported inclusion of the GEP in standard practice at this time. However, this committee is also creating guidelines, not unbreakable rules, and the test is available on the market. Guidelines are intended to be just that, a recommendation by experts for most cases and because they are intended to be applied broadly, it is usual that they are conservative with a high bar for data, and in keeping with this philosophy this committee did not recommend routine use of this test for any melanoma.
Dr. Schwarzenberger: How should GEP fit into our practice?
Dr. Leachman: There are certainly cases where this test is useful, and there are other experts in the field who feel that it is time for the test to begin transitioning into clinical practice. It is unclear what bar is required before transition to standard practice begins to happen — this is a “competing good” for patients, a balance between having enough assurance that the test is not leading to anything unsafe and the assurance that we are pushing the field forward in a positive way.
When technology is in the transition period between being a research-only test and a standard-of-care test, there is an accompanying period of uncertainty about its appropriate use. In this transition period, when it is available on the market, but has not yet been incorporated into formal guidelines, the “art of medicine” needs to be applied and every patient’s case needs to be assessed individually. It is not appropriate for the test to be used on everyone, nor is it appropriate to withhold the test from someone who might benefit from it. As the test transitions further into the clinical realm, the most appropriate population(s) of patients to receive the test will become more apparent, larger data sets will be analyzed, costs will likely come down, and rational guidelines will be set for use.
Dr. Schwarzenberger: Do you anticipate that GEP will eventually transition into regular clinical practice?
Dr. Leachman: An analogy might be the transition from horse and buggy to the automobile. Certainly, the horse was a reliable means of transportation and the auto was less proven, less affordable, and perhaps even less reliable in the beginning. But the auto was faster and didn’t require as much food/labor to keep healthy, and costs came down so that it was affordable for a large percentage of the population. I predict the GEP test will transition in a very similar way. Through use in the larger community, its true utility and the appropriate populations in which it should be used will emerge.
Dr. Schwarzenberger: During this transition period, are there certain cases and situations in which GEP could be helpful in providing additional information?
Dr. Leachman: GEP doesn’t override gold-standard practices, but it superimposes an additional knowledge that can be used to inform difficult and/or equivocal cases. Knowledge is power in these cases, for both the provider and patient. The knowledge may be helpful whether the test is positive or negative as well. For example, a positive (class 2) test may not yet come with enough certainty about outcome to be able to use it to recommend adjuvant therapy, but it can help increase the frequency and intensity of surveillance so that a recurrence will have a greater likelihood of being caught earlier when the tumor burden is less and more treatable.
Similarly, though most thin melanomas are negative (class 1), there are some highly anxious or knowledge-driven individuals who would benefit from that reassurance. In my opinion, what is most vital during the transition period is to have rational, data-based debate among experts with differing opinions, because this offers the greatest chance for successful implementation of a useful test in the right people at the right time. It’s okay that we don’t agree, it means that there is not yet a clear right or wrong answer about the use of the test, but — like every new technology successfully implemented in medicine — we are navigating through the uncertainties as we go.
Dr. Schwarzenberger: What should we do with the results? Our present standard of care does not mandate evaluation beyond serial skin and node exams in patients with early-stage disease. Should we consider ordering imaging on a regular basis for patients deemed high risk by GEP?
Dr. Leachman: My standard practice is to use the test results as a piece of information that guides care of the patient (i.e., this is not a test that is associated with something I feel the patient must do). If a patient has a thin melanoma that would otherwise not be sent for a sentinel lymph node biopsy, but has a class 2 GEP, I would discuss a possible sentinel lymph node biopsy (or refer for that discussion). If the sentinel lymph node biopsy was not performed or was negative, I would work with surgical oncology to follow the patient more intensely (joint follow-up with surgical oncology, decreased threshold for imaging studies, or biopsy, etc.). If the sentinel lymph node was positive, this patient would then have the option of adjuvant therapy.
If a patient with a thicker melanoma that receives a sentinel lymph node biopsy has a class 1 test result and a negative sentinel node biopsy, the surveillance can be decreased, and it is appropriate that they are followed primarily in dermatology. In terms of the imaging I would consider, it would be driven by the symptoms of the patient and would not change in the type of study, just the threshold I have for requesting it. Ultimately, if the patient reports symptoms, it is a “judgement call” regarding whether the sign or symptom is concerning enough to warrant ordering a study (e.g., fine-needle aspiration, ultrasound, MRI, or PET-CT). My threshold for a sign or symptom triggering the ordering of such a study is much lower if the patient has a class 2 test.
Dr. Schwarzenberger: Should we allow GEP information to influence whether a patient gets a sentinel lymph node biopsy?
Dr. Leachman: Yes and no. At this point, we do not know enough about the overlap between the tests to know that one test is sufficient. Some of the decision regarding whether to order the GEP test depends on whether the GEP or the SLN biopsy is done first. If the SLN is positive, then the GEP can superimpose additional prognostic information on that and may influence a patient or provider one way or another about whether to offer adjuvant therapy to a given patient. The bottom line is that the GEP just adds additional information for consideration in the decision-making process, it doesn’t mandate an action. As it becomes more mainstream, that may change and having a class 2 result may be considered reliable enough and high-risk enough to guide the recommendations for adjuvant therapy.
Dr. Schwarzenberger: What is the future of GEP?
Dr. Leachman: I think that these tests are going to become more and more mainstream and will become an integral part of our practice to guide screening and care. Like the automobile, once there is an accepted test, the competition will roll in and we will benefit from better, more efficient, and less expensive tests. Right now, we must accept and tolerate the evolution we are going through to get there and use our best judgement for each patient. If we do, we will all benefit.
Dr. Leachman is professor and chair of the dermatology department at Oregon Health and Science University and serves as director of the Melanoma Research Program at the Knight Cancer Institute. The article appeared in JAAD: doi.org/10.1016/j.jaad.2018.07.028.
