2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Approximately 85% of patients with newly diagnosed melanoma present with what appears to be clinically localized disease, making them candidates for standard treatment with surgery.
Practical Considerations in Metastatic Melanoma
Mario Sznol, MD
Professor, Medicine
Leader, Clinical Research Program in Melanoma
Co-director, Yale SPORE in Skin Cancer
Yale Cancer Center
New Haven, CT
Robert H.I. Andtbacka, MD
Associate Professor, Surgical Oncology
University of Utah School of Medicine
Surgeon, Huntsman Cancer Institute
Salt Lake City, UT
Omid Hamid, MD
Chief, Translational Research and Immunotherapy
Director, Melanoma Therapeutics
The Angeles Clinic and Research Institute
Los Angeles, CA.
Merrick I. Ross, MD
Professor, Surgery
Chief, Melanoma Section
The University of Texas MD Anderson Cancer Center
Houston, TX
Jeffrey A. Sosman, MD
Professor, Medicine
Director, Melanoma and Tumor Immunotherapy Program
Co-leader, VICC Signal Transduction & Cell Proliferation Research Program
Vanderbilt-Ingram Cancer Center
Nashville, TN
Jeffrey S. Weber, MD, PhD
Director
Donald A. Adam Comprehensive Melanoma Research Center
H. Lee Moffitt Cancer Center and Research Institute
Tampa, FL
Approximately 85% of patients with newly diagnosed melanoma present with what appears to be clinically localized disease, making them candidates for standard treatment with surgery. But some of these patients may be harboring microscopic metastases in their lymph nodes.
How best to identify patients who may have microscopic metastatic disease, and should thus be tested via sentinel lymph node biopsy, was among the topics discussed during a recent OncLive Peer Exchange® roundtable discussion entitled “Practical Considerations in Metastatic Melanoma.” Mario Sznol, MD, served as moderator.
In patients found to have a positive node, a biopsy can help “optimize [the patient’s] chance for cure by treating lymph node disease early, and also optimize the chance for long-term regional control within the lymph node basin,” noted panel member Merrick I. Ross, MD.
The panel members also considered whether patients found to have a positive node should go on to receive completion lymph node dissection, and explored the prospect of biomarkers that might eventually be used to guide treatment decisions.
Determining When Biopsy is Appropriate
Determinations about when a patient should undergo sentinel node biopsy are based mainly on the thickness of the lesion, panel members said.
“It used to be 1 mm in Breslow thickness that we would perform this on,” said Robert H.I. Andtbacka, MD. “However, we have lately started looking at the risk of having micro-metastatic disease in those lymph nodes, and we have said that a risk of about 5% is the point at which we would offer the procedure to the patient. That risk of 5% falls at a Breslow thickness of about 0.75 mm, so at our institution, we would do it at 0.75 mm. At other institutions, they still do it at 1 mm.
“We also do look at other parameters of that primary melanoma, such as ulceration and increased mitotic count,” Andtbacka continued. “That’s a mitotic count of 1 per mm2 or greater. We do know that those factors increase the risk of having nodal involvement.”
Mario Sznol, MD
According to Ross, the value in doing a sentinel node biopsy is that metastatic disease, if found, is at a very early stage. That makes early intervention possible, which is why such patients often have better outcomes than those who develop palpable disease after a wide excision, he said. Ross asserted that those whose metastasis is found while microscopic fare 20% better than those whose disease is found when macroscopic, but panelists debated that.
Jeffrey S. Weber, MD, PhD, pointed out that about 80% of patients with melanoma in the MSLT-1 [Multicenter Selective Lymphadenectomy Trial]1 who underwent sentinel node biopsy had negative results, thus diluting that 20% outcome advantage down to 5% for the overall group of melanoma patients with suspected microscopic metastases.
“Then, the question is, given the cost of doing the sentinel node biopsy, is it justified by the modest benefit?” Weber asked. “My personal feeling is probably yes, meaning I can sell it to patients, but I gather there are those who think it’s probably not worth it because there isn’t enough survival.”
Medical oncologist Jeffrey A. Sosman, MD, said there is another advantage to conducting sentinel node biopsy in eligible patients.
“I think there’s a benefit—not an outcome benefit, but a benefit for the patient—in knowing what their risk is,” he said. “The sentinel node gives those patients a real idea of what their ultimate [chance] is of surviving and living a long life, and I think a negative sentinel node can be an incredibly reassuring finding. It’s not 100%, but certainly it makes a big difference.”
Looking ahead, panelists said, the research community could make biopsy decisions easier by continuing to investigate biomarkers that indicate the presence of a positive node, and also by conducting a more definitive trial of the outcomes of patients who undergo biopsies and are found to have microscopically positive nodes.
MSLT-II Explores More Extensive Dissection
These questions are being evaluated now in the MSLT-II trial,2 in which node-positive patients with melanoma who have undergone sentinel lymphadenectomy are randomized to either complete lymphadenectomy or observation with nodal ultrasound.
The MLST-II trial could put to rest the question of whether patients with a positive node should undergo a complete lymph node dissection, Andtbacka said.
“When we find a sentinel lymph node, the majority of patients only have one sentinel lymph node that has micro-metastatic disease in it and the tumor burden tends to be 1 mm or less, so a very small amount of tumor burden,” he said. “The question then becomes that, traditionally, we have recommended that every patient that has a positive sentinel node should have a completion lymph node dissection. The MSLT-II study is going to randomize these patients to either have a completion lymph node dissection versus being followed with ultrasound, the hypothesis being that the 10-year survival in those two randomized patient groups is the same.”
However, since results from that trial won’t be in for several years, “I think it’s important to recognize that, currently, the standard of care for patients that do have micro-metastatic disease in their lymph node basin is to undergo a completion lymph node dissection,” said Andtbacka.
Ross added that he would expect about half the patients with positive biopsies in the trial to have only one positive node, meaning that they may gain all the therapeutic benefit they need simply by undergoing sentinel node biopsy. A problem with the trial, he said, is that those patients will not be treated as a separate cohort.
“Be that as it may,” he said, “I think it’s an important way to identify patients where we don’t have to have as much surgery and we can minimize the morbidity.”
Ross added that conducting completion dissection early could be very important for longer-term regional control of metastatic melanoma.
“When a lymph node dissection or a formal lymphadenectomy is performed for patients with grossly positive nodes, the risk of recurrence in that same nodal basin is pretty high; it’s averaging around 21%, but can be as high as 50% for patients with multiple positive nodes with extracapsular extension or in the head and neck location,” he said. “But the incidence of recurrence in that basin after a completion dissection for microscopic nodal involvement is less than 5%. So we’re improving their quality of life and their outcomes from a regional control perspective, not just from a survival perspective.”
Doing a completion dissection when disease is micro-metastatic, rather than macro-metastatic, is also wise because it leads to less lymphedema and morbidity, noted Andtbacka.
However, Sznol suggested that there could be a down side to completion dissection in this population.
“I’m not 100% sure that the completion lymph node dissection will ultimately be good for the patient,” he said. “There may be a reason to leave those nodes in because they may be important in an immune response against cancer that may be important in preventing relapse. But that’s a topic for another time.”
Considering Molecular Testing
Returning to the subject of biomarkers, the panelists considered whether BRAF mutation status should affect the treatment choices or prognosis of these patients.
“In this patient population that has primary melanoma that has no evidence of nodal involvement, and also in the stage III patients with micro-metastatic disease in the lymph nodes, we would not routinely get that testing, since we would not have necessarily an adjuvant trial for them,” said Andtbacka.
Weber, however, suggested that testing mutational status early on might help with disease management later.
“I think, in the future, every stage III patient should be tested, so that you have the information in the bank,” he said. “You often may have no actionable manifestations, but given that there are two active clinical trials, and there will probably be more trials for stage III resected patients, I think it’s appropriate that any stage III patient have their tumor sent for BRAF testing—and God knows, in a year or two, it will be BRAF and NRAS. It will be 20 genes, and then it will be perhaps even sequenced completely.”
Ross mentioned that a 31-gene array has been studied as a prognostic measure in melanoma.3
“It’s kind of an offshoot of the gene analysis that was done for uveal melanoma, but they’ve extended it to cutaneous melanoma,” he said. “It’s got 28 tumor-specific genes and three control genes. It’s an RTPCR [reverse transcription polymerase chain reaction], so you do it on paraffin-embedded tissue, and it identifies or dichotomizes patients into what’s called class 1 and class 2, and the class 1’s have a very good prognosis and class 2 not as good.”
He noted, however, that “this particular analysis is probably not very good for the stage III patients, because most of those are class 2’s anyway.”
Panelist Omid Hamid, MD, cautioned that a number of questions must be answered before the assay is put into use.
“You have a stage I patient with a class 2 signature,” he said. “Do you image them more often? Do you have an adjuvant therapeutic that you can offer them that can change their course? Do you create more anxiety or less anxiety?”
Although there are many unsettled questions, Hamid and his fellow panelists agreed that the assay is a step in the right direction.
References