PHOENIX, Ariz.—In patients with intermediate thickness localized melanoma, wide excision surgery is usually curative, but metastasis to regional lymph nodes can occur. Some clinicians advocate immediate elective lymphadenectomy in these patients who have positive sentinel node biopsies as a way to improve tumor staging and survival.
But there is disagreement on how much influence this approach has on survival outcomes, especially in patients with only a small tumor burden in the sentinel node. And completion lymphadenectomy exposes patients to the risk of complications, including wound infections and limb edema. In a session at the 2009 Society of Surgical Oncology meeting, two experts debated the merits of completion lymphadenectomy or completion lymph node dissection (CLND) after melanoma metastasis to sentinel lymph nodes (SLN).
CLND helps to accurately stage patients with a positive SLN biopsy, and assists clinicians in making recommendations for adjuvant therapy, according to Merrick I. Ross, MD, professor of surgery and chief of the melanoma section at Houston’ s M.D. Anderson Cancer Center. It also provides locoregional control of metastases, optimizes the chance for a cure, and reduces morbidity, Dr. Ross said.
Taking the opposing side, Alexander M.M. Eggermont, MD, PhD, argued that in patients with a tumor load of less than 0.1 mm in the sentinel node, CLND does not improve survival and can result in unnecessary complications. Dr. Eggermont is head of the department of surgical oncology at the Daniel Den Hoed Cancer Center and professor of surgical oncology at Erasmus University Medical Center in Rotterdam, the Netherlands.
From microscopic to macroscopic
There is an assumption that SLN biopsy has no therapeutic value and is only a staging procedure, Dr. Ross said. But “many of us are seeing patients who develop clinical nodal disease aft er a positive SLN, and they are going on to adjuvant therapy, somewhat defeating the purpose of doing the SLN biopsy from the outset,” he said. “The number of positive nodes is an independent predictor of survival for patients who have stage III disease, and we need to know how many lymph nodes are involved to assess prognosis.”
He also noted that data from the Multicenter Selective Lymphadenectomy Trial showed that patients with nodal metastases who received immediate CLND had a higher five-year survival rate than those in whom CLND was delayed (N Engl J Med 355:307-1317, 2006). Studies also indicate a survival advantage for node-positive patients from CLND; about 20% experience improved survival from the elective procedure, Dr. Ross said.
Deciding whether a patient should be a candidate for CLND becomes a more complicated question when the sentinel node shows only microscopic disease, Dr. Ross acknowledged.
“But it’s pretty rational over time to assume that microscopic disease will lead to macroscopic disease. If you let metastases sit in a lymph node over time, it will become clinically apparent,” he said. This observation is supported by data in the NEJM study, which showed that disease progression occurred in patients with tumor-involved nodes when they underwent surveillance only, he said.
Neck dissection is generally well tolerated, and morbidity rates from CLND are low, Dr. Ross said. Yet it may be right to use a selective approach in low-risk patients who have lymph node metastases in order to avoid unnecessary complications, he acknowledged.
He noted that his group published a study of 359 melanoma patients with positive sentinel node biopsies, and the researchers were able to stratify patients according to the risk of finding additional positive nodes. Using multivariate analysis, they found that tumor thickness, increasing sentinel node burden, and the number of nodes harvested were indicators of low or high risk for finding disease in nonsentinel nodes.
Using these three factors, they developed a model of four quartiles of risk for finding metastases in nonsentinel nodes (J Clin Oncol 26:4296-4303, 2008). Half of the patients with positive sentinel nodes in the JCO study fell into the lowest risk group; less than 4% of these patients were found to have additional positive nodes.
“It may be good to selectively look at excluding node dissection in these low-risk patients. However, that needs to be validated in a larger subset,” Dr. Ross said. “But it’s my feeling that completion of node dissection for positive sentinel nodes is the standard of care, and omission should only occur in a randomized clinical trial or an IRB informed consent prospective registry for low-risk patients.”
Relying on Rotterdam Classification
Dr. Eggermont argued that the incidence of positive nodes in the CLND comes in at only 2.5% of patients with submicrometastases in the sentinel node. “This would suggest that CLND is not necessary,” he told Oncology News International.
He noted that a number of studies have shown that survival rates in these patients are almost identical to those with negative sentinel nodes. He also said that the patients have similar rates of positive nonsentinel nodes compared with patients whose sentinel nodes were initially negative.
Using the Rotterdam Classification system of sentinel node tumor burden could allow oncologists to exclude CLND in patients who have metastases of less than 0.1 mm or submicrometastases.
“If you adhere to these criteria, you look for the largest lesion in any of the sentinel nodes, and that’s the lesion that counts and identifies the score,” he said.
Dr. Eggermont cited a multicenter study of 388 melanoma patients with positive sentinel nodes and primary melanoma to illustrate his argument. The study found that those with submicrometastases of less than 0.1 mm had a prognosis and survival rates similar to those published for sentinel node negative patients (Ann Surg 248:949- 955, 2008).
Metastases of less than 0.1 mm also did not affect the rate of additional positive nodes. “In our multivariate analysis, in this classification system, only sentinel node tumor burden had a significant impact on disease-free survival, distant metastases, and overall survival,” he said.
He noted, however, that using a cutoff of 0.2 mm submicrometastases to exclude CLND would not work as it does for breast cancer patients. In the 2008 study, patients with less than 0.2 mm metastases in the sentinel node experienced a 10% rate of additional positive nodes.
Dr. Eggermont questioned whether the biology of submicrometastases (< 0.1 mm) may be different from that of larger tumor burdens. He noted that these patients might be more accurately classified as false positive. In fact, these patients might be considered sentinel node negative and should not be classified as stage III, he said. They are also highly unlikely to benefit from completion lymph node dissection.
“Completion lymph node dissection is not necessarily indicated in patients where you find submicrometastases,” he said.
“However, I agree that further study in a larger data set is needed.”
“I hope to present that study at ASCO 2009 with updated data on 600 sentinel node positive cases,” Dr. Eggermont said. “Th e study supports our point that the biology of submicrometastases is truly different.”